Introduction: Immune evasion of diffuse large B‐cell lymphoma

(DLBCL) is one of the reasons accounting for its unsatisfactory thera-

peutic effects. Investigating the underlying mechanisms may provide

novel promising potential targets for DLBCL treatment. Exosomes are

known to play a signiﬁcant role in cancer initiation and progression.

However, its role in DLBCL immune escape remains to be elucidated.

Methods: The proportion, immunophenotype and cytotoxic function

of natural killer (NK) cells in DLBCL patients were detected by ﬂow

cytometry. Exosomes were isolated from the culture medium of

DLBCL cells and plasma of DLBCL patients by ultracentrifugation and

then were characterized by immunoblotting, NanoSight and trans-

mission electron microscopy. The role of DLBCL‐derived exosomes on

NK cells was assessed by immunoﬂuorescence and ﬂow cytometry.

Elisa, co‐culture experiments and immunoblotting were performed to

explore the molecular mechanisms of exosomes on NK cells.

Results: We reported the existence of NK cell exhaustion and

increased expression of T‐cell immunoglobulin mucin receptor 3

(TIM3) on NK cells in DLBCL patients. Exosomes were identiﬁed in

DLBCL patient plasma and NK cell functions was negatively associ-

ated with immunosuppressive molecules levels in exosomes. And we

found that exosomes derived from DLBCL cells carried immunosup-

pressive cytokine such as transforming growth factor‐β (TGF‐β) and

lead to decreased percentage and function of NK cells. Finally, we

showed that treatment of NK cells with TGF‐β neutralizing antibody

blocked the effect of exosomes on NK cells.

Conclusions: Altogether, our ﬁndings reveal the prognostic value of

plasma exosome of DLBCL patients, provide a molecular basis

for the role of exosome in immune evasion of DLBCL and highlight the

unique function of exosome as an attractive therapeutic target.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

300 |INTESTINAL T‐CELL LYMPHOMAS: MOLECULAR

INTEGRATIVE ANALYSIS RECOGNIZES DIFFERENT

THERAPEUTIC TARGETS FOR EACH SUBTYPE

R. Manso

, M. Rodriguez

, C. Chamizo

, N. Pérez

, R. Alonso‐Alonso

P. A. Minguez

, J. Borregon

, E. Baez‐Duran

, E. M. Casas‐del Pozo

M. A. Piris

, F. Rojo

, S. M. Rodriguez‐Pinilla

Fundación Jiménez Díaz, Pathology, Madrid, Spain,

Fundacion Jimenez

Diaz, Genetics and Genomics Department and Bioinformatics Unit,

Madrid, Spain

More frequent types of Primary intestinal T‐cell lymphomas are

Enteropathy‐associated T‐cell lymphoma (EATL) and monomorphic

epitheliotropic intestinal T‐cell lymphoma (MEITL). Although they

differ in epidemiology, morphology and immunophenotype they both

shared an unfavourable outcome with poor response to standard

chemotherapeutic regimens. Deﬁnition of the genetic landscape of

these tumours the data is still in process. Here we have analyzed a

series of both EATL and MEITL cases in an attempt to identify tumour‐

speciﬁc therapeutic targets. The series included 7 (EATL) and 9 MEITL

formalin‐ﬁxed parafﬁn‐embedded (FFPE) cases. Both expression and

mutational studies were done by different platforms. GEP analysis

disclosed that the CD40‐pathway was statistically differentially

expressed in the EATL cases, due to the overexpression of BTG2,

HTRA1 or RUNX1 genes; moreover, TNFRSF8 was upregulated in this

group. While RPS6KA3, HDAC2 and FH were highly upregulated in the

MEILT subgroup. Other potential targets differentially expressed in

the EATL subgroup were CD274, TGFB1, SMAD3 or KDR while MEITL

cases displayed increased expression of a set of cytotoxic markers

(GNLY, NKG7 or KLRB1). A total of 24 different genes were found

mutated in the whole series of cases. The number of mutations per case

as well as the allele frequency of most of the mutations was higher in

the MEITL subgroup of tumours. Most EATL patients showed muta-

tions in DNA repair genes, followed by alterations in genes of the

NOTCH, VEGF or PI3K/AKT signaling pathways. On the contrary,

MEITL cases analysed showed mutations in SETD2 gene, the RAS gene

family or the JAK/STAT pathway. Integrative molecular analysis con-

ﬁrms that EATL and MEITL are strongly different disorders, linking

both expression signatures and mutational data. This study suggests

tumor type‐speciﬁc therapeutic targets including immunotherapy for

EATL and CD30, MAPK, JAK/STAT pathways inhibitors for MEITL.

Keywords: Extranodal non‐Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

301 |EXPRESSION OF HUMAN LEUCOCYTE ANTIGENS

GENETIC POLYMORPHISMS IN A CASE SERIES OF FIVE PATIENTS

WITH SEQUENTIAL LYMPHOMAS

E. Verrou

, A. Fylaktou

, I. Diamanti

, K. Tsirou

, T. Triantafyllou

A. Sevastoudi

, Z. Tsinaris

, G. I. Grigoriadou

, P. Konstantinidou

G. Gioula

, E. Katodritou

Theagenio Cancer Hospital, Hematology, Thessaloniki, Greece,

Hippokration Hospital, National Peripheral Histocompatibility Center,

Thessaloniki, Greece,

Theagenio Cancer Hospital, Biochemisty &

Microbiology Department, Thessaloniki, Greece,

Medical School, AUTH,

Microbiology, Thessaloniki, Greece

Introduction: Sequential lymphomas involving both classical Hodgkin

Lymphoma (cHL) and Primary Mediastinal B‐cell Lymphoma have been

previously reported but initial diagnosis of cHL followed by a distinct

diagnosis of Diffuse Large B‐Cell Lymphoma (DLBCL) or vice versa is an

extremely rare phenomenon. Multiple etiological factors are involved

in lymphoma pathogenesis. The major histocompatibility complex

(MHC) has been studied extensively about rising susceptibility in many

diseases and certain different human leukocyte antigen (HLA) associ-

ations with HL and DLBCL have been proposed. This is the ﬁrst report

of HLA investigation in patients with sequential lymphomas.

Methods: We deﬁned HLA class I (HLA‐A, HLA‐B, HLA‐C) and class II

(HLA‐DRB1, HLA‐DQB1) in 5 (2 female/ 3 male) patients with

sequential diagnosis of cHL and DLBCL or vice versa. HLA typing for

396

SUPPLEMENT ABSTRACTS

HLA‐A*, B*, C*, DRB1*, DQB1* was performed using Sequence‐Speciﬁc

Oligonucleotide HLA typing (SSO) and Sequence‐Speciﬁc Primer HLA

typing (SSP). The results of typing were compared using χ

to those

found by 250 healthy adult donors from the bone marrow bank of

Northern Greece.

Results: Three of our patients were HLA‐DRB1*11 (DR11)/HLA‐

DQB1* 03 (DQ7) positive (5/10 haplotypes). The high frequency of

HLA‐DR11/DQ7 combination among our patients (50%) differed not

signiﬁcantly from the observed frequency of this combination among

healthy individuals (21%) (p =0.04 but after using Yate's correction

p=0.09). Τwo of our patients displaying this HLA combination were

treated with ABVD for cHD and developed DLBCL nine and three

years after diagnosis of cHL respectively. Both patients are alive and

in remission for both lymphomas. The third HLADR11/DQ7+patient

was treated with CHOP for DLBCL and developed cHL 21 years after

diagnosis of primary DLBCL. Interestingly while being in remission

for both lymphomas she additionally developed mycosis fungoides

displaying ﬁnally diagnoses of three sequential lymphomas.

Conclusions: Although not reaching statistical signiﬁcance due to

small number of cases our results suggest that HLA analysis may be a

ﬁeld of investigational interest regarding patients with sequential

lymphomas. HLA ‐DR11 has been previously reported as a risk factor

for cHL irrespectively from EBV status. Importantly HLA‐DQ7 is

involved in the pathogenesis of autoimmune diseases. Interestingly

HLA‐DR11/DQ7 combination has been previously reported in asso-

ciation with vulvar lichen sclerosus and asthma. Of note this combi-

nation was also described in one case of familial systemic sclerosis

following exposure to organic solvent. In conclusion our results

emphasize the necessity for collecting and investigating sequential

lymphomas in order to elucidate the underlying contributory mecha-

nisms in the pathogenesis of lymphoproliferative diseases.

Keywords: Lymphoid Cancers – Other, Pathology and Classiﬁcation

of Lymphomas

No conﬂicts of interests pertinent to the abstract.

302 |IMMUNOHISTOCHEMICAL AND HISTOPATHOLOGICAL

STUDY FOR THE DIAGNOSIS AND CLASSIFICATION OF

LYMPHOMAS.

J. J. Jiménez Galainena

, J. Lamadrid García

, Z. Gutiérrez Aleaga

M. Amigó de Quesada

, M. Nazario Dolz

, L. Heredia Manzano

E. Gámez Pérez

, W. Marcial Martínez

, A. Columbié Molina

E. Gracia Medina

, T. González Madruga

, J. Peraza Bordao

National Oncology Institute, Anatomic Pathology Department, Havana

City, Cuba,

National Oncology Institute, Department of Clinical

Oncology, Havana City, Cuba,

National Oncology Institute, Department

of Clinical Oncology, Havana City, Cuba,

Institute of Tropical Medicine

IPK, Department of Anatomic Pathology, Havana City, Cuba

Introduction: The histopathological classiﬁcation and subclassiﬁ-

cation of lymphomas in hematopathology is the gold standard for the

diagnosis of these diseases. Different techniques, such as immuno-

histochemistry, ﬂuorescent in situ hybridization and molecular

studies have been included in the diagnostic ﬁeld of Anatomic Pa-

thology. The information obtained with these techniques combined

with conventional light microscopy study allows a speciﬁc morpho-

logical diagnosis according to the WHO Classiﬁcation permitting the

best treatment in the patients.

Objective: The study with immunohistochemical techniques was to

classify and subclassify these neoplastic lymphoid processes.

Material and methods: The study included 2053 biopsies of different

sites diagnosed as lymphomas at the Pathology Department of the

National Oncology Institute, Havana City, Cuba from the period of

2007 to 2019 with a panel of different antibodies for hematolymphoid

processes employing the technique of immunohistochemistry in

parafﬁn blocks. The manual and automated immunohistochemical

staining were used. The visualization systems included the Polymer

Detection System and the Ultraview Universal DAB Detection Kit.

Results: The Non‐Hodgkin´s B Cell Lymphomas were the predominant

immunophenotypic type (1478 cases)(71.9%).The B Cell Lymphomas

CD20 positive were 1311 cases (63.8 %).The Diffuse Large B Cell

Lymphomas were the most frequent histopathological type (725 cases)

(35.3%).The Diffuse Large B Cell Lymphoma molecular subtypes clas-

siﬁcation (Hans Algorithm)(CD10/Bcl6/Mum1) by immunohisto-

chemistry were the following in the cases studied (96 cases)(2015 ‐

2019): GBC (29 cases)(30.2%), ABC (45 cases)(46.8%),non GBC / non

ABC (18 cases)(18.7%), Double expressor (4 cases)(4.16%).The Pe-

ripheral T Cell Lymphomas (81 cases) and Anaplastic Large Cell Lym-

phoma(67 cases) were the principal histologies diagnosed in the group

of Non‐Hodgkin´s T Cell Lymphomas (153 cases)(7.45%).The extra-

nodal sites were 941 cases (45.8%) and the Head and Neck location

was the common site diagnosed in this group (195 cases)(20.7%).

Conclusion: The study showed the different frequencies, locations

and histological types of the lymphomas in our daily work of a period

of time of 13 years in the Immunohistochemistry Laboratory at the

Pathology Department.

Keywords: GBC (Germinal center B‐cell‐like), ABC (Activated B‐cell‐

like), non GBC / non ABC (non Germinal center B‐cell‐like) / (non

Activated B‐cell‐like)

Keywords: Pathology and Classiﬁcation of Lymphomas

No conﬂicts of interests pertinent to the abstract.

CHRONIC LYMPHOCYTIC LEUKEMIA

303 |PROGNOSTIC VALUE OF APOLIPOPROTEIN A IN

CHRONIC LYMPHOCYTIC LEUKEMIA

X. Yun

, Y. Zhang

, X. Sun

, X. Hu

, H. Zhang

, Z. Yin

, X. Zhang

, M.

Liu

, X. Wang

Shandong Provincial Hospital, Cheeloo College of Medicine, Shandong

University, Department of Hematology, Jinan, Shandong, China

SUPPLEMENT ABSTRACTS

397

Introduction: Lipid metabolism is related to lymphomagenesis and a

potential target for novel therapies of some hematologic malig-

nancies. Chronic lymphocytic leukemia (CLL) cells, like adipose cells

rather than B‐lymphocytes or other leukemia cells, utilize free fatty

acids to produce energy. There is a pronounced drop of high‐

density lipoprotein (HDL) in CLL patients 3‐4 years before the

diagnosis. Apolipoprotein A, the major structural protein of HDL,

takes effect on lipid transportation and immunity. Nevertheless, the

role of ApoA in CLL has not been explored. This study aimed to

investigate the potential role of ApoA in the prognosis of CLL

patients.

Methods: A total of 150 newly diagnosed CLL patients from January

2010 to December 2019 in Shandong Provincial Hospital were

enrolled in this study. Kaplan‐Meier curves and univariate and

multivariate Cox proportion hazard regression models were per-

formed to evaluate the prognostic value of ApoA. Pearson tests,

Spearman tests and Chi‐square tests were performed to evaluate the

correlation between ApoA and other clinical and laboratory

parameters.

Results: Kaplan‐Meier curves revealed that increased ApoA was

associated with longer overall survival (OS) (χ

=14.985; p<0.001;

shown in Figure 1A) and progression‐free survival (PFS) (χ

FIGURE 1 Kaplan‐Meier curves of Apolipoprotein A (ApoA) of overall survival (OS) and progression‐free survival (PFS) of chronic

lymphocytic leukemia (CLL) patients. (A) Increased ApoA was associated with longer OS of CLL patients. (B) Increased ApoA was longer PFS of

CLL patients. (C) Increased AopA post‐therapy was associated with longer OS of CLL patients. (D) Increased AopA post‐therapy was

associated with longer PFS of CLL patients

398

SUPPLEMENT ABSTRACTS

16.190; p<0.001; shown in Figure 1B). Univariate and multivariate

Cox regression revealed that ApoA was a signiﬁcant independent

prognostic biomarker of OS (p=0.028) and PFS (p<0.001). The

level of ApoA did not change with treatment. However, ApoA

levels of CLL patients after therapy were also correlated to OS (χ

6.732; p=0.009; shown in Figure 1C) and PFS (χ

=8.822;

p=0.003; shown in Figure 1D). The low level of ApoA was related

to the advanced stages (p<0.001) and increased β2‐microglobulin

(r=‐0.447, p<0.001), which was known poor prognostic factors

in CLL.

Conclusions: Our study reveals that ApoA is a favorable prognostic

biomarker of OS and PFS in CLL for the ﬁrst time. The novel prog-

nostic biomarker may increase prognostic accuracy of CLL.

The research was funded by: National Natural Science Foundation

(No.82000195, No.82070203, No.81770210, No.81473486 and

No.81270598); Key Research and Development Program of Shandong

Province (No.2018CXGC1213); Technology Development Projects of

Shandong Province (No.2017GSF18189); Translational Research

Grant of NCRCH (No.2021WWB02, No.2020ZKMB01); Shandong

Provincial Natural Science Foundation (No.ZR2020QH094); Taishan

Scholars Program of Shandong Province; Shandong Provincial Engi-

neering Research Center of Lymphoma; Academic Promotion Pro-

gramme of Shandong First Medical University (No. 2019QL018;

No.2020RC007); Technology Development Project of Jinan City

(No.200027182); Shandong Provincial Hospital Youth Talent Plan;

Shandong Provincial Hospital Research Incubation Fund.

Keywords: Diagnostic and Prognostic Biomarkers, Chronic Lympho-

cytic Leukemia (CLL)

No conﬂicts of interests pertinent to the abstract.

304 |TIME TO HEMATOLOGIC RECOVERY PREDICTS

SURVIVAL IN CHRONIC LYMPHOCYTIC LEUKEMIA PATIENTS

TREATED WITH FLUDARABIN, CYCLOPHOSPHAMIDE AND

RITUXIMAB – 11 YEARS OF REAL‐WORLD EXPERIENCE

A. V. Mata

, I. P. Mesquita

, D. Alves

, J. Medeiros

, B. Polo

, C.

Lopes

, M. J. Costa

, C. Martins

, G. Esteves

, J. F. Lacerda

, J.

Raposo

Hospital Santa Maria, Centro Hospitalar Universitário Lisboa Norte,

Hematology and Bone Marrow Transplant Department, Lisbon, Portugal

Introduction: Chemotherapy‐free regimens are frequently the ﬁrst

choice in chronic lymphocytic leukemia/small lymphocytic lym-

phoma (CLL/SLL) treatment. Despite some concerns about short

and long‐term toxicity, chemotherapy is still a highly valuable

option in a speciﬁc subset of patients. We wanted to assess efﬁ-

cacy, overall and progression free survival (OS and PFS), short and

long‐term toxicity and predictors of response in the real‐life

setting.

Methods: Unicentric, retrospective study of CLL/SLL patients treated

with ﬂudarabin, cyclophosphamide, rituximab (FCR) from July 2008

until April 2019.

Results: Sixty‐nine patients were included, 74% were male, median

age of 58 year (IQR 50‐63). Seventy‐eight percent were treated in

ﬁrst line setting, 22% had relapsed/refractory (R/R) disease. Before

treatment, FISH analysis was performed in 67 patients: 4.5% had

(del)17p, 14.9% had del(11q), 37.3% had (del)13q, 28.4% had

triss12 and 29.9% had normal cytogenetic analysis. We found 2/45

patients with TP53 mutation and 19/39 patients had mutated

IgHV.

Complete response (CR) was reached in 41.8%, CR with incom-

plete marrow recovery in 34.3%, partial response in 13.4%, stable

and progressive disease in 3% and 7.5%, respectively. Minimal re-

sidual disease (MRD) was negative in 19/33 patients.

Median follow‐up of 59.6 months, 1st line median PFS of 55.3

months and median OS not reached, with 80.2% of patients alive at

median follow up time. On the R/R setting, median PFS and OS were

41.5 and 58.3 months respectively. Negative MRD had a positive

impact on survival (median PFS not reached vs. 31.3 M, p =0.015), as

well as cytogenetic analysis (p =0.003).

Half of the patients received 6 cycles and more than 75% received at

least 4 cycles. Dose reduction was needed in 18% of the patients, and

36% had some delay in treatment schedule. Hematologic toxicity

grade ≥3 was found in 62%: anaemia 4.3%, neutropenia 49%,

thrombocytopenia 24.6%.

Nearly half had complete hematologic recovery (49.7%) with median

time to recovery of 124 days. Of note, PFS was longer in patients

who took more than 124 days for hematologic recovery (median PFS

110.9 vs 42.7 months, p <0.001; Fig 1).

Infeccious adverse events grade ≥3 were found in 7.5% of patients,

and 17.4% of opportunistic infections (epstein‐barr, herpes zoster, P.

jiroveci). Second neoplasms were found in 13% of patients: 2 acute

FIGURE 1 Time to hematologic recovery (HR) predicts

progression free survival

SUPPLEMENT ABSTRACTS

399

leukemias, 3 adenocarcinomas and 4 skin neoplasms. There were 20

fatalities: 9 for progressive disease (median time to death 49

months), 3 for infectious adverse events (17‐20 days after treat-

ment), 3 for 2

neoplasms and 5 patients for non‐related causes.

Conclusions: In the real‐world setting with a long follow up time, FRC

attained 76% of CR in 1st line therapy, with long lasting remissions

and an acceptable toxicity proﬁle. Cytogenetic analysis, negative

MRD and, interestingly, longer time to hematologic recovery had a

positive impact in PFS.

EA – previously submitted to regional or national meetings (up to

1000 attendees).

Keywords: Diagnostic and Prognostic Biomarkers, Late Effects in

Lymphoma Survivors, Chronic Lymphocytic Leukemia (CLL)

No conﬂicts of interests pertinent to the abstract.

305 |FACILITATED IMMUNOGLOBULIN (fSCIG) USE IN

PATIENTS WITH SECONDARY IMMUNODEFICIENCY DISEASES:

TWO‐YEAR INTERIM RESULTS FROM THE FIGARO STUDY

M. Dimou

, D. Huscher

, C. Hermann

, D. Pittrow

, M. Speletas

M. Borte

National & Kapodistrian University of Athens Medical School, First

Department of Propaedeutic Internal Medicine, Athens, Greece,

Charité‐

Universitätsmedizin Berlin, Institute of Biometry and Clinical

Epidemiology, Berlin, Germany,

Baxalta Innovations GmbH, Global

Medical Affairs, Vienna, Austria,

Technical University of Dresden,

Institute for Clinical Pharmacology, Medical Faculty, Dresden, Germany,

University of Thessaly, School of Health Sciences Faculty of Medicine,

Department of Immunology and Histocompatibility, Thessaly, Greece,

St. Georg Hospital, Hospital for Children and Adolescents, Leipzig,

Germany

Introduction: Secondary immunodeﬁciency diseases (SID) commonly

affect patients with hematological malignancies such as chronic

lymphocytic leukemia (CLL) and lymphoma. Immunoglobulin

replacement therapy (IGRT) is an important therapeutic option for

preventing severe, recurrent, or persistent infections in patients with

hematologic malignancies and SID. Facilitated subcutaneous immu-

noglobulin (fSCIG) is a dual‐vial unit of immunoglobulin G (IgG) 10%

and recombinant human hyaluronidase (rHuPH20) approved in the

European Union as an IGRT for patients with SID. rHuPH20 de-

polymerizes hyaluronan in the extracellular matrix to transiently in-

crease local subcutaneous tissue permeability, allowing for fSCIG

infusion at rates, volumes, and frequencies comparable to intrave-

nous immunoglobulin (IVIG) with better systemic tolerability. This

study assesses fSCIG usage and tolerability over time in patients with

SID seen in routine clinical practice.

Methods: FIGARO (NCT03054181) is a European multicenter,

prospective, observational study. Patients were eligible if they

received ≥1 fSCIG infusion for primary immunodeﬁciency diseases

or SID and provided informed consent. Patients are being followed

for up to 3 years. An interim analysis (data cut‐off date: February

2, 2021) was conducted to evaluate fSCIG usage and tolerability

over the ﬁrst 2 years of follow up in the subgroup of patients with

SID.

Results: Of the 156 patients enrolled, 31 (mean [range] age: 61

[3–88] y) had SID and are included in the baseline (inclusion visit)

analysis. At interim analysis, 1‐and 2‐year follow‐up data were

available for 23 and 12 patients, respectively. Indications for IGRT

were CLL (n =20), non‐Hodgkin lymphoma (n =6), SID due to

immunosuppressive therapy (n =3), and Hodgkin lymphoma (n =2).

At the inclusion visit, fSCIG (median total monthly dose: 30.0 g) was

most recently infused at home (61%) or at a hospital (36%) by the

patient (58%) or a nurse (42%). Most patients received fSCIG

every 4 weeks (68%), infused into a single site. The median

maximum infusion rate was consistent over the 2‐year follow

up: 290 mL/h at inclusion, 268 mL/h at 1 year, and 300 mL/h at 2

years. The median infusion volume was 300 mL across all visits to 2

years, with all patients receiving the full dose as planned. Two pa-

tients each reported 1 adverse drug reaction (ADR; erythema

and headache) at the inclusion visit, with no ADRs reported at

any of the follow‐up visits. One patient changed to IVIG by 12

months.

Conclusions: In patients with SID primarily due to hematologic ma-

lignancies, fSCIG was tolerated over 2‐years' follow up in routine

clinical practice and offered the ﬂexibility of administration at home

or in a hospital setting, either by the patient or a nurse. Infusion

parameters remained consistent over time.

EA – previously submitted to EHA 2021.

The research was funded by: Baxalta Innovations GmbH, a Takeda

company, funded this study. Baxalta US Inc., a Takeda company,

funded writing support.

Keywords: Immunotherapy, Ongoing Trials

Conﬂicts of interests pertinent to the abstract

Dör. Huscher

Educational grants: has received travel compensation from Actelion

Pharmaceuticals, Switzerland, Boehringer‐Ingelheim, Germany, and

Shire

C. Hermann

Employment or leadership position: an employee of the Baxalta In-

novations GmbH, a member of the Takeda group of companies

Stock ownership: a Takeda stock owner

D. Pittrow

Consultant or advisory role: has acted as consultant for Baxalta

Other remuneration: reports personal fees from Actelion, Bayer,

Aspen, Boehringer Ingelheim, Sanoﬁ, Biogen, and MSD

400

SUPPLEMENT ABSTRACTS

M. Speletas

Research funding: institution has received a research grant from Shire

M. Borte

Consultant or advisory role: has participated in advisory boards for

CSL Behring, Octapharma, and Shire

Research funding: institution has received research grant support

from CSL Behring, Octapharma, and Baxalta

306 |EFFECTIVENESS OF THERAPY WITH FLUDARABINE,

CYCLOPHOSPHAMIDE AND RITUXIMAB (FCR) FOR PATIENTS

WITH CHRONIC LYMPHOCYTIC LEUKEMIA

I. Kogarko

, A.K. Golenkov

, E.V. Kataeva

, B.S. Kogarko

, I.I. Ganeev

Institute of Chemical Physics, RAS, Chembio Processes, Moscow, Russian

Federation,

M.F. Vladimirsky Moscow Regional Research and Clinical

Institute (MONIKI), Immunology and Hematology, Moscow, Russian

Federation

Goals: an evaluation of effectiveness of Fludarabine, Cyclo-

phosphomide and Rituximab (FCR therapy) for patients with newly

diagnosed Chronic Lymphocytic Leukemia (CLL) for validation of

results from multi‐centered clinical trials in real clinical practice.

Materials and methods: sixty six (66) patients with newly diagnosis

of CLL were included to this prospective study with average age of

58 years old (39‐73 years old). Stage I ‐II according to Rai staging

system was found in 76% of patients; Stage III‐IV – in 24 % of pa-

tients. All patients received 6‐8 cycles of inductive FCR therapy:

Rituximab 375 mg/m

iv – ﬁrst cycle, followed by 500 mg/m

iv;

Fludarabine – 25 mg/m

iv – days 2,3 and 4 of a cycle; Cyclophos-

phamide – 250 mg/m

,iv– days 2,3 and 4 of a cycle). Patients were

divided into two groups based on duration of induction cycles (IC).

Group 1 had average IC of 36 days; Group 2 ‐49 days. Evaluation of

immunochemotherapy effectiveness was determined by presence of

remissions (IWCLL‐2008), progression‐free survival (PFS) and overall

survival (OS) (Kaplan – Meier method).

Results: Complete Remission (CR) was observed more often in Group

1 (FCR‐36) than in Group 2 (FCR‐49) ‐56.6% and 37.9% of patients,

respectively (p >0.05). Median PFS was 36 months in all patients; 45

months – in Group 1 (FCR‐36) and was not achieved in Group 2

(FCR‐49). Median OS was 45 months in Group 1 (FCR‐36) and was

not achieved in Group 2 (FCR‐49).

Conclusions: conﬁrmation of FCR therapy effectiveness for newly

diagnosed CLL patients was established in real clinical practice

characterized by non‐selective patient groups and prolonged induc-

tion therapy cycles. Results of this study provide validation for re-

sults from multi‐centered clinical trials and could be reproduced on a

greater scale in clinical practice.

Keywords: Chronic Lymphocytic Leukemia (CLL), Chemotherapy,

Immunotherapy

No conﬂicts of interests pertinent to the abstract.

307 |EFFECTIVENESS AND SAFETY OF IBRUTINIB IN

CHRONIC LYMPHOCYTIC LEUKEMIA (CLL) AND MANTLE CELL

LYMPHOMA (MCL) IN BELGIAN ROUTINE CLINICAL PRACTICE

WITH A 3‐YEAR FOLLOW‐UP

A. Janssens

, S. Snauwaert

, D. Bron

, Z. Berneman

, F. Offner

C. Van Bogaert

, B. De Beleyr

, A. Smet

, L. Nielsen

, R. Wapenaar

M. André

Universitair Ziekenhuis Leuven, Department of Hematology,

Leuven, Belgium,

AZ Sint‐Jan Brugge, Department of Hematology,

Brugge, Belgium,

Institut Jules Bordet (ULB), Department of

Hematology, Brussels, Belgium,

Universitair Ziekenhuis Antwerpen,

Department of Hematology, Edegem, Belgium,

Universitair Ziekenhuis

Gent, Department of Hematology, Gent, Belgium,

Janssen‐Cilag NV,

Medical Affairs, Beerse, Belgium,

Janssen‐Cilag NV, Medical Affairs,

Beerse, Belgium,

Janssen‐Cilag NV, Market Access, Beerse, Belgium,

Janssen‐Cilag NV, Medical Affairs, Beerse, Belgium,

Janssen‐Cilag

BV, Statistics Department, Breda, Netherlands,

Université Catho-

lique de Louvain, CHU UCL Namur, Department of Hematology, Yvoir,

Belgium

Introduction: The Belgian ibrutinib Real‐World (RW) Data (BiRD)

study is a multicenter, observational study of adults with CLL, MCL, or

Waldenström's macroglobulinemia. The RW study aims to assess

ibrutinib outcomes (primary: progression‐free survival [PFS] and

overall response rate [ORR]; secondary: included overall survival [OS]

and safety) in patients (pts) with previously untreated or relapsed or

refractory (R/R) CLL and R/R MCL in Belgium. Here we present results

from the third interim analyses (IA) with a median follow‐up of 34.3

months for the CLL cohort and 35.8 months for the MCL cohort.

Methods: Data were collected both prospectively (pro) and retro-

spectively (ret). Total population results were adjusted for left

truncation.

Results: Effectiveness results are presented in the Table.

In the CLL cohort, 221 pts (pro n =71; ret n =150) were

included in the effectiveness population; 64.3% were male, median

age at ibrutinib initiation was 72 (range, 38‐90) years (y), median time

from diagnosis to ibrutinib initiation was 5.8 y, 27.4% were previously

untreated, 59.1% had del17p and/or TP53 mutation, and 72.5% had

no IGHV mutation. ORR was 90.0% (complete response [CR] 16.7%,

partial response [PR] 51.6%, PR with lymphocytosis 21.7%). Of

226 pts in the CLL safety population, 96.9% had ≥1 treatment‐

emergent adverse event (TEAE) and 52.2% had ≥1 serious TEAE.

TEAEs led to dose reduction, interruption, or discontinuation in

21.2%, 31.9%, and 26.5% of pts, respectively. TEAEs of interest

reported as % (% difference from second IA, after an additional

15.3 mos of follow‐up) were: major bleeding 4.9% (‐0.4%); infection

66.4% (9.3%); hypertension 17.3% (5.1%); atrial ﬁbrillation (AF)

10.6% (3.2%); arthralgia/myalgia 16.8% (4.1%); diarrhea 21.7%

SUPPLEMENT ABSTRACTS

401

(3.2%). Median treatment duration for pts with CLL was 25.5 mos

(19.6/28.6 mos pro/ret), and 135/226 pts (58.7%) remain on

treatment.

In the MCL cohort, 76 pts (pro n =20; ret n =56) were included in

the effectiveness population; 76.3% were male, median age at ibrutinib

initiation was 74 (range, 47‐88) y, and median time from diagnosis to

ibrutinib initiation was 2.8 y. ORR was 93.4% (CR 39.5%, PR 53.9%). Of

76 pts in the MCL safety population, all had ≥1 TEAE and 67.1% had ≥1

serious TEAE. TEAEs led to dose reduction, interruption, or discon-

tinuation in 31.6%, 43.4%, and 26.3% of pts, respectively. TEAEs of

interest reported as % (% difference from second IA, after an additional

15.2 mos of follow‐up) were: major bleeding 7.9% (1.3%); infection

64.5% (6.6%); hypertension 6.6% (0%); AF 21.1% (4.0%); arthralgia/

myalgia 11.8% (0%); diarrhea 35.5% (5.2%). Median treatment dura-

tion for pts with MCL was 24.7 mos (22.8/25.3 mos pro/ret), and 26/76

pts (34.2%) remain on treatment.

Conclusions: Results from the BiRD study conﬁrm the sustained

effectiveness of ibrutinib in both CLL and MCL. TEAEs were

manageable in elderly pts with no new safety signals.

EA – previously submitted to EHA 2021.

The research was funded by: Janssen Pharmaceutica NV, and Phar-

macyclics LLC, an AbbVie Company.

Keywords: Chronic Lymphocytic Leukemia (CLL), Indolent non‐

Hodgkin lymphoma, Molecular Targeted Therapies

Conﬂicts of interests pertinent to the abstract

A. Janssens

Consultant or advisory role: Janssen, Roche, Gilead, Abbvie, Novartis,

Amgen, Sanoﬁ‐Genzyme

Honoraria: Janssen, Roche, Abbvie, Novartis, Amgen, Sanoﬁ‐

Genzyme, Celgene

Research funding: Janssen

Educational grants: Janssen, Celgene, Abbvie, Roche

D. Bron

Consultant or advisory role: Janssen

Honoraria: Janssen

Research funding: Janssen

F. Offner

402

SUPPLEMENT ABSTRACTS

Research funding: Janssen

C. Van Bogaert

Employment or leadership position: Janssen

B. De Beleyr

Employment or leadership position: Janssen

Stock ownership: Janssen, Pﬁzer

A. Smet

Employment or leadership position: Janssen

L. Nielsen

Employment or leadership position: Janssen

R. Wapenaar

Employment or leadership position: Janssen

Stock ownership: J&J

M. André

Consultant or advisory role: Takeda, Bristol‐Myers‐Squibb, Kar-

yopharm, Gilead, Incyte

Research funding: Roche, Johnson & Johnson, Takeda

Educational grants: Roche, Bristol‐Myers‐Squib, Celgene, Gilead,

Abbvie

308 |HEMORRHAGIC MANIFESTATIONS OF LONG TIME

ANTICOAGULANTS TREATMENT IN PATIENTS WITH CHRONIC

LYMPHOCYTIC LEUKEMIA RECIVING IBRUTINIB.

E. Emelina

, G. Gendlin

, I. Nikitin

Pirogov Russian National Research Medical University, Hospital Therapy

№2, Moscow, Russian Federation

Introduction. Long‐term use of anticoagulants (AC) for atrial ﬁbril-

lation (AF) in patients (pts) with chronic lymphocytic leukemia (CLL)

in the setting of ibrutinib‐associated thrombocytopathy (Ib), throm-

bocytopenia and platelet variability is associated with a particular

risk of hemorrhagic hemorrhage complications.

Method. We have observed in dynamics for 5 years 217 pts with CLL

constantly receiving Ib. AF occurred in 39 pts (19.9%) during treatment

Ib, 21 pts had AF before Ib. 46 pts with CLL and AF received AC from 12

to 58 months, of which 19 women aged 70.0 (64.0‐74.0) years and 27

men aged 67.5 (63.5‐70.0) years. 20 pts receive apixaban (api), 20

rivaroxaban (riva), 6 dabigatran (dabi). We tracked all emerging hem-

orrhagic manifestations in pts with CLL, taking AC. During the period of

use of AC, the inﬂuence of platelet level and variability on the occur-

rence of hemorrhages in these pts was evaluated.

Results. In 65.9% of pts with CLL receiving Ib and AC hemorrhages

persist throughout the observation period, represented by nosebleeds

(34.5%), hemorrhage in the sclera of the eye (27.6%), gross hematuria

(20.7%), gingival bleeding (13.8%), hematomas (75.9%), petechiae

(10.3%), bleeding from a rectal tumor in 2 pts, hemorrhage in the

anterior chamber of the eye in 1 pts. Macrohematuria required

discontinuation of AC for up to 2–3 weeks and transfer to the minimum

dose api. In one pts with recurrent macrohematuria AC was canceled.

Hemorrhage in the sclera of the eye required outpatient ophthalmic

treatment without withdrawal of AC. Nosebleeds were recurrent and

led to the transfer of pts to the minimum dose api. Hematomas were

permanent, with predominant localization on the arms and legs, in 1 pts

they appeared on the face, neck, tongue. Several types of hemorrhages

were observed in 47.7% of patients. Transfer to the minimum dose api

2.5 mg * 2 times a day was required in 38.6% of patients due to

recurrent nosebleeds, gross hematuria, combined hemorrhages. Clin-

ically signiﬁcant hemorrhages occurred less frequently in patients

treated with api compared with riva (p =0.019). We did not observe the

inﬂuence of the level and variability of platelets on the occurrence of

hemorrhages in pts with CLL receiving AC. Hemorrhages did not lead

to dose changes and withdrawal of Ib.

Conclusions. With long‐term use of AC in pts with CLL and AF, among

clinically signiﬁcant hemorrhages, recurrent nosebleeds, recurrent

hemorrhages in the sclera of the eye and gross hematuria prevailed,

which led to a change in the tactics of using AC. Due to hemorrhages

38.6% of pts with CLL required transfer to the minimum api dose.

Short‐term discontinuation of AC required pts with gross hematuria.

The level of platelets and their variability in our study did not have a

signiﬁcant effect on the occurrence of hemorrhages. Hemorrhages did

not lead to dose reduction and withdrawal of Ib.

Keywords: Chronic Lymphocytic Leukemia (CLL)

No conﬂicts of interests pertinent to the abstract.

309 |INFLUENCE OF ACTIVE CARDIOMONITORING ON THE

OVERALL SURVIVAL OF PATIENTS WITH CHRONIC

LYMPHOLEUKEMIA RECEIVING IBRUTINIB.

E. Emelina

, G. Gendlin

, I. Nikitin

Pirogov Russian National Research Medical University, Hospital Therapy

№2, Moscow, Russian Federation

Introduction: overall survival of patients (pts) with chronic lympho-

cytic leukemia (CLL) depends not only on the severity of the CLL, the

anticancer therapy used and the comorbidity of pts. The most

important role in the management of these pts is played by the su-

pervision of a cardiologist.

Methods: we examined and observed in dynamics for 5 years 217 pts

with CLL, constantly receiving ibrutinib, which induces arterial hy-

pertension (AH) and atrial ﬁbrillation (AF) in a part of the pts. All pts

underwent echocardiography (ECHO), 24‐hour Holter ECG moni-

toring (HM), assessment of comorbidity using the Charlson index and

screening of fragility using the G8 questionnaire. Daily measurement of

blood pressure, heart rate in the morning and in the evening with

keeping a diary of measurements, was recommended for all pts, but

only 81 pts were performed, who constantly contacted us remotely

using instant messengers and made up an active cardiac monitoring

SUPPLEMENT ABSTRACTS

403

group. Correction of treatment was carried out in accordance with

these data.

Results: a study of the overall survival of patients with CLL receiving

ibrutinib, depending on cardiac monitoring, was carried out, starting

from the ﬁrst visit. The age of pts in the active cardiac monitoring

group (n =81) and in the rest of pts (n =136) did not differ and

amounted to 66.0 (60.0‐70.0) years and 66.0 (59.0‐74.0) years

respectively. The number of men and women in the groups was

comparable. In the active cardiac monitoring group, there were

signiﬁcantly more pts with AH ‐86.5% and with AF ‐42.7%

compared to 50.4% with AH and 15.9% with AF in the remaining pts

(p <0.0001 in both cases) and a comparable number of pts with

coronary artery disease. According to screening HM, there were

more pts with short episodes of AF in the active cardiac monitoring

group ‐31.4% versus 8.0% in the rest of the pts (p <0.0001).

Accordingly, the number of pts receiving cardiac treatment in the

active cardiac monitoring group was 87.6%, the rest was 53.6%

(p <0.0001). ECHO parameters did not differ in the groups. In-

dicators that signiﬁcantly affect survival in the general group

(Charlson index, scores of the G8 questionnaire) did not have signif-

icant differences in the active cardiac monitoring group and in other

pts with CLL. The groups also did not differ in hematological status

and the number of cases of second tumors. Despite a signiﬁcantly

more pronounced cardiac comorbidity, CLL pts under active cardiac

monitoring, including continuous remote monitoring, demonstrated

better survival compared to other patients (p <0.0001).

Conclusions: carrying out active cardiac monitoring, including con-

stant remote observation, allows achieving higher overall survival

rates for CLL pts, despite the more severe cardiac status compared to

other patients under the periodic supervision of a cardiologist.

Keywords: Cancer Health Disparities

No conﬂicts of interests pertinent to the abstract.

INDOLENT LYMPHOMAS

310 |INDUCTION R

FOLLOWED BY MAINTENANCE IN

PATIENTS WITH RELAPSED/REFRACTORY TRANSFORMED

FOLLICULAR LYMPHOMA: INTERIM ANALYSIS FROM THE

PHASE 3B MAGNIFY STUDY

F. Lansigan

, V. Parameswaran

, A. Yacoub

, C. M. Reynolds

S. H. Shao

, S. G. Moore

, M. Coleman

, D. J. Andorsky

, G. S.

Nowakowski

, M. J. Rummel

, J. Li

, J. R. Ahn

, M. Gharibo

J. P. Sharman

Dartmouth‐Hitchcock Medical Center, Division of Hematology and

Oncology, Lebanon, New Hampshire, USA,

Avera Research Institute,

Hematology, Transplantation, and Oncology, Sioux Falls, South

Dakota, USA,

University of Kansas Cancer Center, Department of

Hematology‐Oncology, Westwood, Kansas, USA,

IHA Hematology

Oncology Consultants, Ann Arbor, Ypsilant, Michigan, USA,

Compass

Oncology, Rose Quarter Cancer Center, Portland, Oregon, USA,

UNC REX

Cancer Center, REX Hematology Oncology Associates, Raleigh, North

Carolina, USA,

Clinical Research Alliance Inc, Weill Cornell Medicine,

New York, New York, USA,

Rocky Mountain Cancer Centers, US

Oncology Research, Boulder, Colorado, USA,

Mayo Clinic, Division of

Hematology, Rochester, Minnesota, USA,

Justus‐Liebig Universität,

Med. Clinic IV, Hematology, Giessen, Germany,

Bristol Myers Squibb,

Global Biometrics and Data Sciences, Princeton, New Jersey, USA,

Bristol Myers Squibb, US Medical Affairs Hematology, Princeton, New

Jersey, USA,

Willamette Valley Cancer Institute and Research Center,

US Oncology Research, Eugene, Oregon, USA

Background: The lenalidomide +rituximab (R

) combination has

shown complementary clinical activity and is tolerable in both

untreated and relapsed or refractory (R/R) patients with indolent

non‐Hodgkin lymphoma (iNHL). Few patients with iNHL, ∼2%

per year (Link et al. J Clin Oncol. 2013), experience transformation

to aggressive lymphoma, ie, transformed follicular lymphoma (tFL).

Methods: MAGNIFY is a multicenter, phase 3b trial (NCT01996865)

in patients with R/R FL grades 1—3b, tFL, marginal zone lymphoma,

and mantle cell lymphoma. Lenalidomide 20 mg on d 1—21 of a 28‐d

cycle +rituximab 375 mg/m

/wk cycle 1 and then every 8 wk

starting with cycle 3 (R

) is given for 12 cycles followed by

1:1 randomization in patients with stable disease, partial response, or

complete response/complete response unconﬁrmed (CR/CRu) to R

vs rituximab maintenance for 18 mo. The primary end point is pro-

gression‐free survival (PFS) by 1999 International Working Group

(IWG) criteria. Secondary end points include safety, CR rate, duration

of response, duration of CR (DOCR), time‐to‐response, time‐to‐next

antilymphoma therapy, and overall survival. This analysis evaluates

the interim primary endpoint of overall response rate (ORR) by 1999

IWG criteria and safety of R

induction in patients with tFL in the

induction intention‐to‐treat population.

Results: As of August 28, 2020, 14 patients with tFL were enrolled

(median age, 70.5 y [range, 42—83]); 93% had stage III/IV disease, and

64% had bulky disease (>7 cm or >3 cm �3 lymph nodes). All patients

received prior rituximab‐containing therapy. Median number of prior

antilymphoma treatments was 2 (range, 1—7). Seven patients (50%)

had disease progression from FL ≤24 mo after any prior treatment.

Median time from initial FL to tFL diagnosis was 29.2 mo (range, 0—

212.1). Median follow‐up was 37.1 mo for patients still alive; 8 pa-

tients (57%) have died, 7 from PD and 1 from adverse event (AE; res-

piratory arrest). Median duration of R

treatment during induction was

7.38 mo (range, 1.5—11.3). ORR was 50% (n =7), with 3 patients (21%)

having a CR/CRu. Median DOCR was not reached (95% CI, 2.8—NR),

and median PFS was 7.5 mo (95% CI, 2.6—32.5). The most common

treatment emergent AEs (TEAE) of any grade were (n, %) diarrhea (7,

50%), nausea (7, 50%), neutropenia (7, 50%), constipation (6, 43%),

fatigue (6, 43%), cough (5, 36%), decreased appetite (5, 36%), and hy-

pokalemia (5, 36%). Grade 3/4 TEAEs occurring in >1 patient included

neutropenia (6, 43%), pneumonia (3, 21%), hypokalemia (3, 21%),

leukopenia (2, 14%), and hypertension (2, 14%). Nine patients (64%)

discontinued induction lenalidomide (7 for PD and 2 withdrawals by

patient), and 5 (36%) entered maintenance treatment.

404

SUPPLEMENT ABSTRACTS

Conclusions: R

has activity in tFL. Although there were frequent

early progressions, patients who achieve CR or partial response may

have prolonged responses.

The research was funded by: Bristol Myers Squibb

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma

Conﬂicts of interests pertinent to the abstract

F. Lansigan

Consultant or advisory role: Celgene/BMS

Honoraria: Celgene/BMS

Research funding: Acrotech Pharmaceuticals

A. Yacoub

Other remuneration: SPEAKER'S BUREAU: Incyte

M. Coleman

Stock ownership: immunomedics

Research funding: Abbvie, Bristol Myers, Celgene, Genentech, Gilead,

BeiGene, Innocare, Merck, Pﬁzer, Roche,

D. J. Andorsky

Consultant or advisory role: Abbvie, BMS

Research funding: Celgene, Astra‐Zeneca, Epizyme

G. S. Nowakowski

Consultant or advisory role: Selvita, Debiopharm Group, Kite/Gilead,

Celgene, MorphoSys, Genentech

Research funding: Celgene, NanoString Technologies, MorphoSys

J. Li

Employment or leadership position: BMS

Stock ownership: BMS

Research funding: BMS

Educational grants: BMS

J. R. Ahn

Stock ownership: BMS

M. Gharibo

Stock ownership: BMS

J. P. Sharman

Consultant or advisory role: Abbvie; Acerta Pharma; BMS; Celgene;

Pharmacyclics; TG Therapeutics

Research funding: Acerta Pharma; Celgene; Pharmacyclics; TG

Therapeutics

311 |AUTOLOGOUS HEMATOPOIETIC STEM CELL

TRANSPLANTATION IN FOLLICULAR LYMPHOMA IN THE ERA

OF NOVEL THERAPIES‐A RETROSPECTIVE ANALYSIS BY POLISH

LYMPHOMA RESEARCH GROUP.

A. T. Szlauer‐Stefańska

, W. Sawicki

, E. Paszkiewicz‐Kozik

J. Romejko‐Jarosińska

, T. Czerw

, S. Giebel

Maria Skłodowska‐Curie National Research Institute of

Oncology, Department of Bone Marrow Transplantation and

Onco‐Hematology, Gliwice, Poland,

Military Institute of Medicine,

Department of Internal Medicine and Hematology, Warsaw, Poland,

Maria Skłodowska‐Curie National Research Institute of Oncology

(MSCNRIO) in Warsaw, Department of Lymphoid Malignancies, Warsaw,

Poland

Introduction: Autologous hematopoietic stem cell transplantation

(autoHSCT) in follicular lymphoma (FL) is a therapeutic option

for patients with relapsed or refractory (r/r) disease. With

the universal access to rituximab and the advent of novel ther-

apies, the role of autoHSCT remains to be re‐established.

We have performed a retrospective analysis to identify fac-

tors affecting the outcome of autoHSCT in r/r FL in the rituximab

era.

Methods: The study included 66 consecutive patients (39 women)

with r/r FL and a history of rituximab treatment. Median age was 51

years (range 22‐73). Median number of prior lines of systemic

therapy was 2 (1‐7). Patients underwent autoHSCT in 3 centres

from the Polish Lymphoma Research Group in years 2000‐2019.

Individuals with grade G3b FL or transformation to aggressive

lymphoma were excluded. Complete response was achieved before

autoHSCT in 46 (70%) cases. Most frequently used conditioning

regimens were BEAM (BeEAM): carmustine (or bendamustine),

etoposide, cytarabine, melphalan (n =39, 59%) or 12 Gy TBI‐based

protocols (n =17, 26%).

The analysis of prognostic factors was focused on conditioning

regimens, pre‐transplant rituximab resistance, history of early

(less than 24 months) treatment failure (ETF) and Deauville

scoring of ≥3 in pre‐transplant PET‐CT (data available for 56

patients).

Results: Median number of transplanted CD34+cells was 3.7 �10

kg (0.8‐48.3). All patients engrafted. The median time to neutrophil

recovery was 11 days (9‐16) while platelet count >20 �10

/L was

reported after a median of 11 days (0‐23).

Median follow‐up was 48 months (1‐235). 20 patients (30%) relapsed

during the follow‐up, 6 patients (9%) died. Only one patient died

without disease progression in day +333 due to pulmonary causes.

Median overall survival (OS) was not reached, whereas median

progression‐free survival (PFS) was 115 months. Probabilities of OS

and PFS at 5 years were 85% and 55%, respectively. One case of

secondary malignancy was reported (Hodgkin lymphoma 10 years

after autoHSCT).

In a univariate analysis the probability of OS at 5 years was

decreased for patients with prior ETF (81.8% vs. 100%, p =0.04) and

Deauville ≥3 (83.3% % vs 97.7%, p =0.02), whereas no adverse

factors for PFS were found.

Conclusion: Results of our analysis show that autoHSCT is a safe

and valid option for r/r FL patients with low transplant‐related

mortality. Long‐terms PFS suggests that autoHSCT may have

curative potential in a subset of patients with FL. Prior ETF and

Deauville score of ≥3 in PET‐CT before transplantation are adverse

factors for OS. In these cases, alternative treatment options should

be considered.

SUPPLEMENT ABSTRACTS

405

Keywords: Stem Cell Transplant

No conﬂicts of interests pertinent to the abstract.

312 |A NON‐INTERVENTIONAL STUDY OF OBINUTUZUMAB

IN PATIENTS WITH PREVIOUSLY UNTREATED ADVANCED

FOLLICULAR LYMPHOMA (URBAN): IMPACT OF COVID‐19

PANDEMIC ON ENROLLMENT AND SAFETY

A. Pinto

, P. L. Zinzani

, L. Arcaini

, G. Gritti

, C. Patti

, E. Pennese

S. De Lorenzo

, C. Piparo

, E. Guardalben

, M. Ladetto

Istituto Nazionale Tumori, Fondazione ‘G. Pascale', IRCCS, Ematologia

Oncologica, Naples, Italy,

Università di Bologna, Istituto di Ematologia

“Seràgnoli”, Bologna, Italy,

Fondazione IRCCS Policlinico San Matteo,

Divisione di Ematologia, Pavia, Italy,

Ospedale Papa Giovanni XXIII,

Ematologia, Bergamo, Italy,

Azienda Ospedali Riuniti Villa Soﬁa‐Cervello,

Oncoematologia, Palermo, Italy,

Ospedale Santo Spirito, Dipartimento di

Oncologia ed Ematologia, Pescara, Italy,

Azienda Ospedaliera Rilievo

Nazionale (A.O.R.N.) S. Giuseppe Moscati, Ematologia, Avellino, Italy,

Roche Spa, Clinical Operations, Monza, Italy,

Roche Spa, Medical

Affairs, Monza, Italy,

Azienda Ospedaliera Santi Antonio e Biagio e

Cesare Arrigo, Ematologia, Alessandria, Italy

Background. The introduction of new agents has improved the

prognosis of follicular lymphoma (FL). Unfortunately, COVID‐19

pandemic may have affected disease management.

Objective. The non‐interventional, retrospective/prospective UR-

BAN study is aimed to evaluate effectiveness and safety of

Obinutuzumab‐based treatment in real‐life practice of patients with

untreated advanced FL with FLIPI≥2. The study is ongoing in 46 sites

in Italy. The objective of this interim analysis is to evaluate patient

characteristics, potential changes in the management due to COVID‐

19 pandemic, preliminary efﬁcacy at the end of induction (EOI) and

safety data collected about 1 year after the ﬁrst patient enrolled.

Methods. Participants should have received at least 2 cycles of

Obinutuzumab‐chemotherapy induction before the enrollment

(retrospective part of the study). The study, launched in Sep‐2019

includes 6‐8 induction courses, 2‐year maintenance with Obinutu-

zumab monotherapy, and 1‐year follow‐up period after the last

Obinutuzumab administration. Two time periods have been analyzed:

406

SUPPLEMENT ABSTRACTS

pre‐COVID‐19 phase (until 24‐Feb‐2020) and COVID‐19 phase (25‐

Feb‐2020 to 09‐Oct‐2020; clinical cut‐off date).

Results. Overall, 266 patients (median age 61 years, 56% females)

were enrolled: 186 patients began the treatment pre‐COVID‐19

(pre‐C group) and 80 patients during COVID‐19 (C group).

Among the trends, not statistically signiﬁcant, from this analysis,

more patients in the C group vs pre‐C group initiated treatment with

B‐symptoms (26.2% vs 19.4%; p =0.357) and more advanced (III‐IV)

disease stage (95% vs 90.9%; p =0.605).

Fewer patients in the C group than in the pre‐C one (41.2% vs

48.9%; p =0.309) received bendamustine, resulting in reversed rates

for CHOP adoption (55% in C and 44.1% in pre‐C; p =0.133).

As to safety, in the retrospective part of the study neutropenia

occurred in 22.9% of patients, infusion‐related reactions in 12.8%,

thrombocytopenia in 9.4%, and infections in 7.5%. In the prospective

part they were 17.7%, 0.8%, 2.3%, and 8.6%, respectively. One pa-

tient had a grade 3 COVID‐19 infection, which required treatment

withdrawal. No data is yet mature about possible changes in main-

tenance strategies due to COVID‐19.

Preliminary efﬁcacy data, available for 164 patients at EOI, show

that 134 patients (81.7%) achieved a PET‐CT CR (conﬁrmed by BM

evaluation for 46/73 patients positive at baseline), and 25 (15.2%) a

PR.

Conclusions. The interim analysis of the URBAN study shows that

Obinutuzumab‐based treatment is associated with a good safety and

efﬁcacy proﬁle. Furthermore, the observed trends might suggest that

during pandemic the adoption of a watch‐and‐wait strategy is wider

and the use of bendamustine reduced, possibly due to concerns on

more severe and prolonged induced lymphodepletion; therefore

COVID‐19 might have inﬂuenced the clinical approach to the treat-

ment of FL patients.

The research was funded by: Roche Spa, Monza, Italy

Keywords: Indolent non‐Hodgkin lymphoma

Conﬂicts of interests pertinent to the abstract

C. Piparo

Employment or leadership position: Roche employee

E. Guardalben

Employment or leadership position: Roche employee

313 |CLINICAL CHARACTERISTICS AND OUTCOMES OF

WALDENSTRÖM MACROGLOBULINEMIA PATIENTS, TREATED

AND FOLLOWED IN A PUBLIC BRAZILIAN CENTER.

G. B. Dufﬂes Amarante

, R. S. Szor

, A. Duran

, L. B. d. O. Alves

G. Martinez

, J. Pereira

, V. G. Rocha

Serviço de Hematologia, Hemoterapia e Terapia celular, Hospital das

Clínicas, Faculdade de medicina da Universidade de São Paulo (USP),

Hematology, São Paulo, Brazil,

Laboratory of medical investigation in

pathogenesis and directed therapy in onco‐immuno‐hematology,

Faculdade de medicina da Universidade de São Paulo (USP), Hematology,

São Paulo, Brazil

Waldenström macroglobulinemia (WM) is a rare disease, account-

ing for less than 2% of non‐Hodgkin`s lymphoma. Current data on

the clinical and epidemiological characteristics of the disease are

widely from European and American centers. Data on Brazilian

WM patients are scarce in the literature, represented only by case

reports.

This study aimed to retrospectively assess clinical features,

treatment options and outcomes of patients diagnosed with WM

from 1999 to 2017 at a single university center in Sao Paulo,

Brazil.

Fifty patients met the eligibility criteria, 31 (62%) males with

median age at diagnosis of 63.5 years (range 42‐84). The median

(IQR) values of blood tests at presentation were: hemoglobin 9.1g/dl

(7.9‐11), platelet 178.000/mm

(112‐240), β

M 3.6g/dl (2.6‐5.0),

calcium 9.1mg/dl (8.8‐9.6), IgM 3495mg/dl (368‐9759). Median time

from diagnosis to start of treatment was 41 days (min‐max, 0‐966).

IPSSWM‐R was very low and low risk in 38%, while 42% of cases

were high or very high risk. All but 3 patients (watch and wait) were

treated. The following indications for treatment were available in 38

patients: anemia (71%), thrombocytopenia (26%), B symptoms (47%),

hyperviscosity syndrome (18%), progressive adenopathy/bulky dis-

ease (8%), symptomatic neuropathy (8%) and hemolytic anemia (8%).

The main chemotherapy protocols for ﬁrst‐line therapy were: CVP

(37.5%), chlorambucil (20.8%) and FC (10.4%). Overall response rate

(ORR) was 45.7%, mainly partial remission (PR) (37.1%). Stable dis-

ease occurred in 28.5% and progressive disease in 25.7% of cases.

Thirty‐one patients (62%) received second‐line therapy. The most

prescribed protocols were CVP (n =10), CHOP (n =5), Fludarabine

(n =4), chlorambucil (n =4) and FC (n =3). PR was the commonest

response (45%). The median follow‐up time was 4.0 years (0‐10, IQR).

Estimated 4‐year OS was 63% (CI 95%, 46‐76), and 4‐year PFS was

35% (CI 95%, 21‐49).

Our study reﬂects a real‐world experience of a rare disease in a

public health university center in a middle‐income country, with

resource constrained limitations. Our study has limitations, mainly

related to its retrospective design. Moreover, genotype test is not

reimbursed and there is no approval for Rituximab, Bendamustine or

Bruton tyrosine kinase inhibitors in Brazil's public health system for

WM. Even with those limitations, our results seem to be comparable

with previous reported data of the DRC protocol, albeit with lower

ORR. Our patients were younger, with lower hemoglobin levels and

with more B symptoms than other studies. In addition, difﬁcult access

to health care assistance delays diagnosis. Efforts should be made to

perform genotype tests and make available speciﬁc treatment for

MW. Medical education, establishment of a reference center and a

national registry may improve knowledge of MW and outcomes of

patients in Brazil and other countries.

Keywords: Indolent non‐Hodgkin lymphoma, Multiple Myeloma

No conﬂicts of interests pertinent to the abstract.

SUPPLEMENT ABSTRACTS

407

314 |ENDPOINTS IN MARGINAL ZONE LYMPHOMAS: A

SYSTEMATIC REVIEW REVEALS WIDE HETEROGENEITY ACROSS

TRIALS AND CALLS FOR STANDARDIZATION.

C. Bommier

, M. Ruggiu

, A. Monegier

, E. Zucca

, C. Thieblemont

, J.

Lambert

Hôpital Saint Louis, Onco‐hematology department, U1153 Inserm ‐

ECSTRRA Team, Paris, France,

Hôpital Saint Louis, Hematopoietic stem

cell transplant, Paris, France,

Hôpital Saint Louis, Statistics and Bioin-

formatics, Paris, France,

Ospedale Regionale di Bellinzona e Valli, IELSG,

Bellinzona, Switzerland,

Hôpital Saint Louis, Hemato‐oncology, Paris,

France,

Hôpital Saint Louis, Statistics and Bioinformatics, Inserm U1153

ECSTRRA Team, Paris, France

*the two last authors contributed equally to this work.

Introduction:Marginal zone lymphoma (MZL) is a heterogeneous

disease, which indolent course requires long and costly trials to

evaluate novel therapeutics. There is a need to ﬁnd more speciﬁc and

shorter endpoints. Our aim was to carry out an inventory of the

endpoints used in trials involving at least 1 MZL patient.

Methods:We performed a systematic review of the endpoints used in

published and registered trials in marginal zone lymphoma. We

searched via PubMed, The Cochrane Library, clinicaltrials.gov and

clinicaltrialsregister.eu for published and registered “clinical trials”

using the keyword “marginal zone lymphoma”. We included studies

on human adult patients (18 years or older) treated for a MZL; we

excluded studies focusing on pediatric populations, cutaneous mar-

ginal zone lymphoma and on use of allogenic stem cell transplant.

Inclusions were double‐checked and data extraction was performed

by two blinded reviewers.

Results:1192 references were identiﬁed by the initial screening.

Among the 309 included references (111 published, 198 registered),

214 (69%) were phase II, 65 (21%) phase III and 30 (10%) phase I/II

trials. The majority of them were open‐label (n =295, 95%) non‐

randomized (n =256, 83%) trials, concerned all subtypes of MZLs at

once (n =239, 77%), and were often merged with non‐MZL patients

(n =231, 75%). Overall/complete response rate (ORR/CRR) was the

most used primary endpoint (n =208, 67%), followed by progression‐

free survival (PFS, n =49, 16%). The most frequent secondary end-

points were overall survival (OS, n =153, 50%), PFS (n =142, 46%) and

408

SUPPLEMENT ABSTRACTS

ORR/CRR (n =116, 38%). Distribution of endpoints was similar when

considering trials with only MZL patients. ORR/CRR was signiﬁcantly

more used as primary endpoint in phase 2 trials (70.5% vs 19.4%),

while PFS was more used in phase 3 trials (58.1% vs 10.1%, p <0.001).

Choice of primary endpoint was not inﬂuenced by neither family of

treatment (targeted vs no targeted therapy, p =0.92) nor funder type

(p =0.46). Same endpoints were deﬁned differently across published

trials. ORR was reported either as best response (76%) or response at

a ﬁxed time point (24%); death was not mentioned as a component of

PFS in 13/97 (13%) trials; 9 different deﬁnitions were used for Event‐

free survival and as many for Time to failure.

Conclusion: Trials involving MZL patients showed marked heteroge-

neity both in the choice and deﬁnitions of primary and secondary

endpoints. Standardization is needed among international in-

vestigators in order to allow comparability between trials.

Primary and secondary endpoints used in trials involving MZL pa-

tients, according to their study design. (up) involving at least 1 MZL

patient; (bottom) including only MZL patients. MZL: marginal zone

lymphoma. Phase III trials are colored in green, phase II trials in yellow

and phase I/II trials in pink.

The research was funded by: First author of this abstract is funded by

a public grant from Agence Régionale de Santé Île‐de‐France as part

of a PhD graduation.

Keywords: Indolent non‐Hodgkin lymphoma, Therapeutics and Clin-

ical Trials in Lymphoma ‐Other

No conﬂicts of interests pertinent to the abstract.

315 |CLINICAL OUTCOMES IN PATIENTS RELAPSED/

REFRACTORY AFTER ≥2 PRIOR LINES OF THERAPY FOR

FOLLICULAR LYMPHOMA: A SYSTEMATIC LITERATURE REVIEW

AND META‐ANALYSIS

S. Kanters

, B. Kahl

, A. Wiesinger

, B. Gurung

, E. H. Limbrick‐

Oldﬁeld

, A. Sudhindra

, J. Thornton Snider

, A. R. Patel

RainCity Analytics, Vancouver, Canada,

Washington University School

of Medicine in St. Louis, 2. Oncology Division, Department of Medicine, St

Louis, USA,

Kite, A Gilead Company, Santa Monica, USA,

Health

Economics & Outcomes Research, IQVIA, London, UK

Introduction: Patients with follicular lymphoma (FL) can have high

response rates to early lines of treatment. However, among FL pa-

tients relapsed/refractory (r/r) after ≥2 prior lines of therapies (LOT),

remission tends to be shorter and there is limited treatment guid-

ance. This study sought to evaluate the clinical outcomes for r/r FL

after ≥2 prior LOT using available treatment options.

Methods: Electronic databases were searched on 2

June 2020.

Eligible studies were comparative or non‐comparative interventional

or observational studies of systemic therapies among adults with FL r/

r after ≥2 prior LOT. Prior LOT must have included anti‐CD20

monoclonal antibodies and alkylating agents, in combination or

separately. Overall response rate (ORR) and complete response (CR)

were analyzed using inverse‐variance weighting with Freeman‐Tukey

double‐arcsine transformations. Kaplan‐Meier (KM) analysis for

progression‐free survival (PFS) and overall survival (OS) were con-

ducted using digitized curves and the Guyot algorithm. Survival ana-

lyses were stratiﬁed by ≥2 prior LOT and ≥3 prior LOT groups and

were restricted to the observational cohorts as a sensitivity analysis.

Results: The meta‐analysis included 12 studies published from 2014‐

2020. Sample size for response outcomes was 340, for survival

outcomes with ≥2 prior LOT was 1024 and for ≥3 prior LOT was

502. The estimated ORR in the ≥2 prior LOT group was 56.0% (95%

conﬁdence interval [CI]: 47.2 – 64.5) and proportion of patients with

CR was 12.2% (95% CI: 8.0 – 17.3). The median OS was 54.4 months

(95% CI: 45.8 – 76.0) and median PFS was 10.3 months (95% CI: 9.3 –

11.1). The 24‐month OS decreased from 66% in the ≥2 prior LOT

group to 60% in the ≥3 prior LOT group (Table 1), with a similar

trend in PFS at 24‐month (28% vs 24%).

Conclusions: This study found that few r/r FL patients with ≥2 prior

LOT achieve CR, and despite some beneﬁt, approximately 1/3 of

patients die within 24 months. The shorter median PFS with

increasing prior LOT suggest treatment durability is suboptimal in

TABLE 1Key outcomes

SUPPLEMENT ABSTRACTS

409

later LOT. These ﬁndings indicate that patients are underserved by

current treatments, demonstrating a need for new treatments that

can achieve high rates of durable response in this disease.

EA – previously submitted to ASCO and EHA 2021.

The research was funded by: This project was funded by Kite, a

Gilead company.

Keywords: Indolent non‐Hodgkin lymphoma

Conﬂicts of interests pertinent to the abstract

S. Kanters

Employment or leadership position: RainCity Analytics

Research funding: RainCity Analytics has received funds from for‐

proﬁt healthcare companies for research

B. Kahl

Consultant or advisory role: Abbvie, Acerta, AstraZeneca, Celege,

Kite, Genentech, Pharmacyclics, MEI

Research funding: Genentech, Celgene, Acerta, AstraZeneca

A. Wiesinger

Employment or leadership position: Kite, A Gilead Company

Stock ownership: Kite, A Gilead Company

B. Gurung

Employment or leadership position: IQVIA Ltd

Research funding: IQVIA Ltd

E. H Limbrick‐Oldﬁeld

Employment or leadership position: RainCity Analytics

A. Sudhindra

Employment or leadership position: Atara Biotherapeutics, Kite

Pharma

Stock ownership: Atara Biotherapeutics, Gilead

J. Thornton Snider

Employment or leadership position: Kite, A Gilead Company

Stock ownership: Gilead Sciences

Research funding: Previously employed at PrecisionHEOR, which

receives research funding from life sciences companies.

A. R Patel

Employment or leadership position: Kite, A Gilead Company

Stock ownership: Kite, A Gilead Company

MANTLE CELL LYMPHOMA

316 |THE CONTRIBUTION OF MONOCYTE‐TO‐PLATELET

RATIO TO PREDICT OVERALL SURVIVAL IN MANTLE CELL

LYMPHOMA: A SINGLE‐CENTER SURVEY

F. Lorenzano

, G. Santuccio

, L. Caruso

, A. Chiarenza

, A. Figuera

G. Motta

, F. Di Raimondo

, A. Romano

University of Catania, Dipartimento di Chirurgia e Specialità Medico‐

Chirurgiche, Catania, Italy,

AOU Policlinico Rodolico San Marco, Divi-

sione di Ematologia, Catania, Italy

In mantle cell lymphoma (MCL) the activity and number of tumour

immune microenvironment (TIME) cells can be deﬁned by some im-

mune dysregulation and chronic inﬂammation markers, such as

neutrophil‐to‐lymphocyte ratio (NLR) and monocyte‐to‐platelet ratio

(MPR). In lack of studies about the contribution of NLR and MPR in

MCL‐TIME lacking, we investigated their prognostic value in a

retrospective single‐center series.

Methods: We reviewed the clinical ﬁles of 122 MCL patients, with a

conﬁrmed histological diagnosis of MCL who accessed our centre

from 21/06/2000 to 28/05/2020. Due to the lack of complete data

regarding some of these patients, the statistical analyses were

conducted on 108 of them. We analysed the correlation between

immune dysregulation markers and overall survival (OS) at 60

months.

Results: The median age at diagnosis was 69 years (ranging between

36 and 85 years); 81 of the patients (74.31%) were male. A complete

blood count was performed routinely for each of these patients to

white blood cell count with differential (median WBC 8580/µL

ranging from 2530/µL to 217790/µL; median ANC 3960/µL ranging

from 520/µL to 14220/µL; median ALC 2000/µL ranging from 60/µL

to 197750/µL; median AMC 560/µL ranging from 20/µL to 11310/

µL), platelet count (median PLT 170500/µL ranging from 19000/µL to

412000/µL), haemoglobin levels (median Hb 12.35 g/dL ranging from

7 g/dL to 16.3 g/dL) and various immune dysregulation and chronic

inﬂammation markers (such as NPR, with a median of 24.6 and

ranging from 4.19 to 277.33; NLR, with a median of 1.7 and ranging

from 0.03 to 111.6; LMR, with a median of 4.17 and ranging from

0.17 to 450.14; PLR, with a median of 0.0893 and ranging from

0.0009 to 1.54).

In the whole cohort of patients, univariate survival analysis showed a

signiﬁcative prolonged OS at 60 months in patients carrying baseline

MPR ≥7.5 (median OS: 35.3, HR =1.82; p =0.03; p value: 0.03). In

patients with age ≥70, only Ki67 ≥25% was correlated with OS

(HR =2.39; p =0.008). In patients with age <70, several factors were

signiﬁcative in univariate analysis, including: presence of B symptoms

(HR =3; p =0.001), PS‐ECOG ≥3 (HR =3.13; p =0.008), male gender

(HR =3.39; p =0.03), Ki67≥25% (HR =2.39; p =0.02), MPR ≥7.5

(HR =2.44; p =0.03) and NLR ≥6 (HR =2.63; p =0.01). In multivariate

analysis, presence of B symptoms, male gender, MPR ≥7.5 and NLR ≥6

were independent predictors of OS at 60 months.

Conclusions: Between the immune dysregulation markers analysed,

MPR and NLR can be considered valid prognostic factors in MCL

setting. NLR, according to our results, is a valuable marker for less

than 70 years old patients, while MPR resulted to be valid in the

whole cohort of patients. Larger and functional studies of mono-

cyte behavior in MCL are required to explore the contribution of

MPR in predicting clinical outcome in this setting.

Keywords: Diagnostic and Prognostic Biomarkers

No conﬂicts of interests pertinent to the abstract.

410

SUPPLEMENT ABSTRACTS

317 |REAL WORLD OUTCOME IN MANTLE CELL LYMPHOMA:

A FRENCH RETROSPECTIVE STUDY IN ELDERLY PATIENTS

BETWEEN 2005 AND 2018

B. Branco

, P. Bories

, L. Ysebaert

, C. Laurent

, B. Cabarrou

, T.

Botin

, Y. Leveneur

, L. Ghenim

, M.‐H. Gaspard

, N. Hess

, W.

Vaillant

, M. Carreiro

, L. Oberic

IUCT‐Oncopole, Haematology, Toulouse, France,

IUCT‐Oncopole, Pa-

thology, Toulouse, France,

IUCT‐Oncopole, Biostatistics, Toulouse,

France,

Centre Hospitalier Intercommunal Castres‐Mazamet, Haematol-

ogy, Castres, France,

Centre Hospitalier de Bigorre, Internal Medicine,

Tarbes, France,

Centre Hospitalier de Rodez, Oncology, Rodez, France,

Clinique Claude Bernard, Oncology, Albi, France,

Clinique du Pont de

Chaume, Oncology, Montauban, France,

Centre Hospitalier de Auch,

Haematology, Auch, France,

Centre Hospitalier de Montauban,

Oncology, Montauban, France

Introduction: First line treatment of patients with mantle cell

lymphoma (MCL) otherwise ineligible to stem cell transplant is

not well standardized. While being predominantly diagnosed in

elderly people, scarse data is available in this population outside

trials. We aimed to describe clinical management in a large unse-

lected cohort of MCL patients, and evaluate their outcomes in a real

world setting.

Methods: We identiﬁed patients aged ≥65 years and diagnosed with

MCL, between 2005‐2018 in a 2.8 M inhabitants region (accounting

for 4.25% of France population). Demographics, baseline biological

and clinical features were analyzed using descriptive statistics.

Progression‐free (PFS) and overall (OS) survival were estimated using

Kaplan‐Meier method.

Results: We identiﬁed 260 patients with a median age at diagnosis of

75 years (range 65‐90), with 71 being ≥80 years old. 83% of patients

had ECOG 0‐1, while only 14% having B symptoms. Median CIRS‐G

comorbidities score was 5 (range 0‐14). 91% patients had stage III‐IV

disease, blastoid variant represented 14% of all diagnoses. MIPI score

was available in 199 patients: high in 66%, intermediate in 29% and

low in 5%.

250 patients received ≥1 treatment immediately after diagnosis

(n =211) or after a period of watch and wait (n =39). Deferred

treatment had no impact neither on OS or PFS. Patients received

anthracyclines (51%), chlorambucil (24%), bendamustine (13.3%) or

cytarabine (5.2%) based regimens, the majority in combination with

rituximab (n =241).

After a median follow‐up of 56.6 months, median OS was 61.3

months (Fig 1A) and median PFS was 23.6 months for the treated

patients population. The 2‐and 5‐y OS rates were 77.2% and 54.0%

respectively, with 2‐and 5‐y PFS rates of 47.4% and 21.4%,

respectively.

Type of ﬁrst‐line chemotherapy was not associated with OS nor PFS,

except for patients receiving chlorambucil whom had worse OS than

those treated by anthracyclines (HR 1.7 ; 95%CI =[1.1;2.6]).There

was no improvement in survival curves based on year of ﬁrst therapy,

2005‐2011 versus 2012‐2018 (Fig 1B).

In multivariable analysis, only CIRS‐G score, medullar involvement

and MIPI score were associated with worse OS.

POD24 (progression of disease within 24 months of treatment) had

an impact on survival (HR 3.5 ; 95%CI =[2.0;6.1] ; p <0.0001),

SUPPLEMENT ABSTRACTS

411

median OS was 50.1 versus 113.7 months for POD24 versus no

POD24 patients. A total of 66 patients received ibrutinib at any line

of relapse, with a median OS and PFS of 16.6 and 8.2 months

respectively, after a median follow‐up of 16.4 months. Upon pro-

gression while on ibrutinib, median OS was only 6.3 months.

Conclusions: This large retrospective study dedicated to elderly pa-

tients with newly diagnosed MCL conﬁrms previously described

prognostic factors. We observe no improvement in survival since

2005 despite availability of new therapeutic options (rituximab

maintenance, bendamustine, ibrutinib). Continuous evaluation of

novel strategies and their incorporation in ﬁrst line treatment to

improve outcomes are needed in this frail population.

Keywords: Indolent non‐Hodgkin lymphoma, Lymphoid Cancers ‐

Other

No conﬂicts of interests pertinent to the abstract.

318 |MANTLE CELL LYMPHOMA, A SINGLE CENTER

EXPERIENCE

A. Roumelioti

, C. Giatra

, M. Dellatola

, E.‐K. Dimitraki

, M.

Bouzani

, G. Kanellis

, D. Ikonopoulou

, K. Kaouranis

, N. El‐Gkotmi

A. Loutsidi

, P. Kosmas

, I. Darmani

, K. Souravla

, V. Babali

, D.

Gardeli

, L. Ligdi

, E. Andreou

, G. Tounta

, Z. Mellios

, F. Kar-

aolidou

, I. Tsonis

, T. Tzenou

, S. Gigantes

, I. Baltadakis

, D. Kar-

akasis

, K. Sakellariou

, D. Rontogianni

, M. Bakiri

, T. Karmiris

Evaggelismos General Hospital, Hematology & Lymphoma Department,

BMTU, Athens, Greece,

Evaggelismos General Hospital, Department of

Hematopathology, Athens, Greece,

Evaggelismos General Hospital,

Department of Pathology, Athens, Greece

Introduction: Mantle Cell Lymphoma (MCL) is an indolent B‐cell

non Hodgkin Lymphoma (NHL) with a worse outcome compared

to other low grade NHL. The prognosis of the disease is dictated by

various biologic and clinical factors. Although MCL is chemo-

sensitive in newly diagnosed patients, soon it relapses. A signiﬁcant

improvement of its outcome has been achieved by the administra-

tion of Rituximab (R) based regimens and the intensiﬁcation of ﬁrst

line treatment by high dose Cytarabine and autologous stem cell

transplantation.

Methods: To reveal the experience of our Centre on the diagnosis,

treatment and outcome of this difﬁcult to treat lymphoma, we

collected data from 51 consecutive patients, diagnosed at our

department between 1999 and 2020.

Results: Thirty nine (39) men and 12 women with median age 68

years (range 33‐88) were diagnosed with MCL. All but one had

typical histology. One patient manifested the blastoid variant. Ki67

was available for 39 biopsy: In 57% specimens Ki67 was >30% and in

43% Ki67 was ≤30%. In 84% of cases the performance status at

diagnosis was good (PS 0‐1) and only 16% showed PS 2‐4. The ma-

jority of patients (90%) had advanced stage disease (III‐IV). High MIPI

score was observed in 59%, intermediate in 18% and low in 22%.

Elevated lactate dehydrogenase was observed in 58% of cases.

Nineteen patients received R ‐Cyclophosphamide, Hydroxydaunor-

ubicin, Oncovin, Prednisone (R‐CHOP), 11 an intensiﬁed regimen, 14

patients R ‐Bendamustin, 5 patients R ‐Chlorambucil and 2 subjects

had no treatment. Response to treatment showed as follow: Com-

plete Response (CR) 52%, Partial Response (PR) 36%, Stable disease

(SD) 2% and Progressive Disease (PD) 10%. Among patients who

received an intensiﬁed treatment, 9 underwent an autologous

transplantation. Twenty two patients with responsive disease bene-

ﬁced of R maintenance. Over a median follow up of 62.8 months

(range 0.3 ‐262), the median overall survival (OS) was 165 months

(range 0.3 ‐262). Twenty three out of 44 (52%) patients with

responsive disease, relapsed. The median progression free survival

(PFS) was 59 months (range 6 ‐140). Sixteen out of 23 (69%) patients

achieved a 2nd response. The treatments administered were: R‐

CHOP (6 patients), R‐Bendamustin (6 patients), platinum based

regimens (3 patients), Ibrutinib (5 patients), corticosteroids (2 pa-

tients), R ‐Chlorambucil (1 patient). The median PFS and OS after the

1st relapse were 26.4 months (range 6.9 ‐165) and 27 months (range

0.2 ‐233) respectively. Response to 1st line treatment (p 0.0001),

lower MIPI score (p 0.001) and intensiﬁed regimen (p 0.0004) were

correlated with a better survival.

Conclusions: Increasing the percentage of 1st complete response by

intensifying the 1st line treatment could improve the survival of

patients with MCL.

Keywords: Indolent non‐Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

319 |IBRUTINIB IN PATIENTS WITH RELAPSED/REFRACTORY

MANTLE CELL LYMPHOMA: REAL LIFE DATA FROM THE 'RETE

EMATOLOGICA DEL LAZIO PER I LINFOMI' (RELLi)

S. Mariani

, S. Pelliccia

, M. P. Bianchi

, A. Di Rocco

, L. Petrucci

E. Maiolo

, R. Battistini

, F. Fanelli

, G. Tomei

, F. Palombi

, E. Papa

L. Pupo

, A. Andriani

, S. Hoaus

, M. Cantonetti

, C. Tesei

, A. Tafuri

A.O.U. Sant'Andrea, Sapienza University of Rome, Dipartimento di

Medicina Clinica e Molecolare, Rome, Italy,

Hematology, Sapienza

University of Rome, Department of Translational and Precision Medicine,

Rome, Italy,

Università Cattolica S. Cuore, Department of Hematology,

Rome, Italy,

AO San Camillo‐Forlanini, Haematology and Haematopoietic

Stem Cells Transplant Unit, Rome, Italy,

Fabrizio Spaziani Hospital, He-

matology, Frosinone, Italy,

IRCCS Regina Elena National Cancer Institute,

Hematology and Stem Cell Transplant Unit, Rome, Italy,

Haematology

Division, University of Tor Vergata, Department of Biomedicine and Pre-

vention, Rome, Italy

Introduction: Treatment of relapse/refractory (R/R) mantle cell

lymphoma (MCL) patients lacks a gold standard therapy. Ibrutinib is a

Bruton Tyrosine Kinase (BTK) inhibitor approved in this setting.

412

SUPPLEMENT ABSTRACTS

Results obtained from the registration studies (PCYC‐1104,

MCL2001, MCL3001) reported 68%, 63% and 72% of overall

response rate (ORR), and 1.9, 10.5 and 14.6 months of median

progression‐free survival (PFS), respectively. However, a less favor-

able outcome is reported in terms of ORR in real life studies: ORR

ranging from 36,4 % to 65% and median PFS from 12,9 to 15 months

(Broccoli A et al, 2018, Epperla N et al, 2017). Therefore, we aimed to

retrospectively evaluate R/R MCL patients treated with Ibrutinib in

RELLi network.

Methods: Patients included in this study were diagnosed with R/R

MCL between January 2004 and April 2020 in several institutions

belonging to the RELLi. Clinical characteristics, treatments and out-

comes were analyzed.

Results: Thirty‐four patients (M:21, F:13) with a median age of 67,5

years (range 42‐84) were included. At diagnosis, 82.3% of patients

presented at Ann Arbor stage III/IV and 76.4% showed a classic

form (11.8% for both blastoid and pleomorphic variant). Patients

with high MIPI score were 70.6% and those with Ki‐67 >30% were

85%. Patients were treated with different frontline therapies:

intensive chemotherapy followed by ASCT, R‐CHOP (or R CHOP‐

like) and Bendamustine‐based regimens (R‐BAC/R‐Benda) in 29.4%,

26.4% and 32.3%, respectively. Patients analyzed were in second,

third and fourth line of treatment, 61%, 23% and 16% respectively.

After Ibrutinib treatment, patients achieved complete remission

(CR), partial response (PR) and progressive disease (PD) in 58%,

18% and 24%, respectively. With respect to age, 41% were <65

years, among these 64% underwent ASCT as frontline treatment.

Older patients (>65 years) were 59%, of whom 45% underwent

treatment based on R‐BAC/Benda and 55% based on R‐CHOP like.

Notably, 38% of patients were >70 years and 14% of them (5

cases) were >80 years. The toxicities were: G1 hyperamylasemia

(one case), G2 neutropenia (two cases) and G1 thrombocytopenia

(one case). One acute myocardial infarction was reported in a pa-

tient in CR requiring Ibrutinib discontinuation. The median overall

survival (OS) was 50% after a median follow‐up of 141 months and

the median PFS was 21 months. The median OS for CR was not

reached, and for PR and PD patients was 11 and 23 months,

respectively (p 0.023).

Conclusions: Our study conﬁrms the efﬁcacy of Ibrutinib in pa-

tients with R/R MCL, with a good safety proﬁle. However, dis-

crepancies emerged with respect to real‐life studies (ORR of 76%

vs. 36,4‐65% and median PFS of 21 months vs. 12.9‐15 months),

conﬁrming in our series data from registration studies. Moreover,

the toxicity data analysis showed no need of dose reduction ac-

cording to age, encouraging this approach also in the elderly R/R

MCL patients.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma

Conﬂicts of interests pertinent to the abstract

A. Tafuri

Research funding: Roche, Novartis, Celgene, AbbVie

AGGRESSIVE NHL

320 |CHIDAMIDE WITH PEL REGIMEN (PREDNISONE,

ETOPOSIDE, LENALIDOMIDE) FOR ELDER OR FRAIL PATIENTS

WITH RELAPSED/REFRACTORY DIFFUSE LARGE BCELL

LYMPHOMA

Y. Wang

, H. Xue

The Afﬁliated Hospital of Qingdao University, Hematology, Qingdao,

China

Background: Diffuse large B‐cell lymphoma (DLBCL) is the most

common form of non‐Hodgkin lymphoma and is increasing in inci-

dence, particularly for elderly people. Treatment for relapsed/re-

fractory DLBCL is evolving rapidly due to emergence of novel drugs,

and HDACi is an important one of these drugs.

Methods: In this study we report the experiences with chidamide

containing regimen CPEL(chidamide, prednisone, etoposide and lena-

lidomide)with or without rituximab(R)in refractory or relapsed DLBCL

(R/R DLBCL) who were not eligible for intensive chemotherapy and

autologous stem cell transplantation(ASCT) due to physical condition

of their age, comorbidities or unwillingness to accept ASCT. This is an

observational retrospective analysis. We retrospectively analyzed the

characteristics of this cohort of patients and their response to the

novel regimen of chidamide in combination with PET regimen (pred-

nisone, etoposide and lenalidomide).

Results: Between June 2018 and June 2020, a total of 34 patients with

R/R DLBCL were treated with CPEL+/‐R at least 2 cycles in the Afﬁl-

iated Hospital of Qingdao University. The interim ORR was 76.5%

(44.1% PR, 32.4%CR), and the ultimate ORR was 50% (20.6% PR, 29.4%

CR), after the median follow‐up of 20.9 ±6.2 months (95% conﬁdence

interval [CI]8.7‐33.1), the median progression‐free survival(PFS) was

13.4 ±3.6 months (95%CI 6.2‐20.6) and the median overall survival

(OS) was19.3 ±3.9 months (95%CI 11.7‐27.0). The 1‐year expected

PFS rate was 50%, and the 1‐year expected OS rate was 68.5%. The

most common grade 3/4 adverse events was hematology,11(32.4%)

patients had neutropenia, 4 patients (11.8%) suffered anemia and 2

patients (5.9%) had thrombocytopenia. 8 patients suffered grade 3/4

non‐hematologic adverse events, among which one patient had grade 4

PE (pulmonary embolism) and other grade 3/4 adverse events were

fatigue, infection, thrombogenesis, anorexia and hyponatremia.

Conclusions: Chidamide‐PET regimen is an efﬁcient and tolerable

regimen for older R/R DLBCL patients unﬁt for intensive

chemotherapy.

The research was funded by: This work was supported by the Beijing

CSCO Clinical Oncology Research Foundation Programs

(Y247HR2018‐270).

Keywords: Molecular Targeted Therapies, Combination Therapies,

Immunotherapy

No conﬂicts of interests pertinent to the abstract.

SUPPLEMENT ABSTRACTS

413

321 |TEMOZOLOMIDE, ETOPOSIDE, DOXIL,

DEXAMETHASONE, IBRUTINIB AND RITUXIMAB FOR THE

TREATMENT OF RELAPSED OR REFRACTORY CENTRAL

NERVOUS SYSTEM LYMPHOMA IN CLINICAL PRACTICE

Author(s): U. Ringeleviciute

, S. Cernauskiene

, A. Zucenka

R. Pileckyte

, L. Griskevicius

Vilnius University Hospital Santaros Klinikos, Hematology, Oncology and

Trasfusion Medicine Center, Vilnius, Lithuania

Introduction. Introduced in 2015, DA‐TEDDI‐R (dose adjusted

temozolomide, etoposide, liposomal doxorubicin, dexamethasone,

ibrutinib and rituximab) regimen has showed acceptable toxicity and

promising survival rates not only in primary CNSL but also in sec-

ondary CNSL which is generally associated with poor prognosis.

Methods. We retrospectively collected data on CNSL patients

treated with TEDDI‐R (Inhibition of B Cell Receptor Signaling by Ibru-

tinib in Primary CNS Lymphoma, June 12, 2017, https://www.cell.com/

cancer‐cell/pdfExtended/S1535‐6108(17)30167‐8) in Vilnius Uni-

versity Hospital Santaros Klinikos from 2018 to 2020. Response was

evaluated by gadolinium‐enhanced MRI according to the Interna-

tional PCNSL workshop response criteria. The end points were

overall survival (OS), progression free survival (PFS) and grade 3‐5

toxicity. The toxicity was evaluated according to Common Termi-

nology Criteria for Adverse Events v5.0.

Results. 13 patients (9 male) were included. The median age was 55

years (28‐70) and median performance score was 2 (0‐4). 12 patients

had relapsed of refractory R/R CNSL and one patient was intolerant to

high‐dose methotrexate. High‐grade B‐cell histology was veriﬁed in all

cases. 7/13 (54 %) cases had primary CNSL. 5 of these patients had had

autologous stem cell transplantation (auto‐SCT) prior to relapse. Of 6

(46%) secondary CNSL patients, high‐dose methotrexate CNS pro-

phylaxis had previously been administered in 5 (83 %). In total, 43

cycles of TEDDI‐R were administered, with a median of 3 cycles per

patient (1‐6). Of 7 primary CNSL patients, 2 had progressive disease

(PD). 5 patients achieved PR, 2 of them proceeded to consolidation

with whole brain radiotherapy. 1 patient remains in PR (14.9 months)

and 1 died due to progressive disease (PD) (17.9 months). Of 3 un-

consolidated patients, 1 remains in PR (21.8 months) and 2 had PD

(21.7 and 12.2 months). Of 6 secondary CNSL patients, 1 patient died

due to pneumonia at day 6 of treatment. Of 5 evaluable patients, 1

patient had progressive disease (PD). 4 patients had achieved PR and

proceeded to auto‐SCT. 2 of them achieved and remain in complete

remission (CR) (34.7 and 19.8 months) and 1 remains in PR (10.2

months). One patient died due to auto‐SCT toxicity. Grade 3‐5 adverse

events occurred in 54% of patients. After a median follow up of 21.7

months (95% CI 14.9‐22.7), median OS was not reached. Among pa-

tients who responded to TEDDI‐R, the median PFS was 21.9 months

(95% CI 13.1‐21.9).

Conclusions. In R/R secondary CNSL, TEDDI‐R may be used as a

bridge to auto‐SCT with curative intent. The regimen may induce

responses in some patients with R/R primary CNSL.

Keywords: Combination Therapies, Aggressive B‐cell non‐Hodgkin

lymphoma

No conﬂicts of interests pertinent to the abstract.

TEDDI‐R outcomes in CNS lymphoma patients (n =13)

414

SUPPLEMENT ABSTRACTS

322 |OUTCOME OF PATIENTS WITH NEWLY DIAGNOSED

PRIMARY CNS LYMPHOMA AFTER HIGH‐DOSE METHOTREXATE

FOLLOWED BY CONSOLIDATION WHOLE‐BRAIN

RADIOTHERAPY AND CYTARABINE

P. Piriyakhuntorn

, T. Rattanathammethee

, S. Hantrakool

, C. Chai‐

Adisaksopha

, E. Rattarittamrong

, A. Tantiworawit

, L.

Norasetthada

Chiang Mai University, Division of Hematology, Department of Internal

Medicine, Faculty of Medicine, Chiang Mai, Thailand

Introduction: Current treatment of primary CNS lymphoma (PCNSL)

is a high‐dose methotrexate (HD‐MTX)‐based polychemotherapy.

The addition of cytarabine to HD‐MTX‐based chemotherapy im-

proves clinical outcomes, however the combination therapy increases

toxicity. Sequential chemotherapy may decrease toxicity without

altering efﬁcacy. This study aimed to analyze efﬁcacy and safety of

sequential chemotherapy and cranial radiation for newly diagnosed

PCNSL patients.

Methods: This was a single center, retrospective cohort study

of consecutive newly diagnosed immunocompetent PCNSL patients

treated with HD‐MTX (5 cycles of 3 g/m

every 2 weeks) followed

by consolidation WBRT (total 46 Gy) and cytarabine (2 cycles of

3 g/m

/d for 2 days every 3 weeks) from January 2013 to

December 2020. Initial WBRT before HD‐MTX was allowed in pa-

tients with Karnofsky performance status less than 70%. Primary

outcome was progression‐free survival (PFS). Key secondary out-

comes were response rate, treatment‐related toxicity, and overall

survival (OS).

Results: Of 41 patients, 25 patients had complete response (CR) and

10 patients had partial response, which inferred to an overall response

rate (ORR) of 85.4% and a CR rate of 60.9%. More than 90% of patients

were able to tolerate and complete HD‐MTX. Twenty‐one patients

(51.2%) received initial WBRT before chemotherapy. The incidence of

≥grade 3 hematologic and non‐hematologic toxicities were 4.8% and

17.1%, respectively. Treatment‐related mortality rate was 2.4%. There

was no difference in toxicity between patients with age <60 and ≥60

years. At the median follow‐up duration of 39.8 months, median PFS

was 35.2 months (95%CI 12.4 – 69.3) and median OS was 46.5 months

(95%CI 21.8 – NR).

Conclusion: High‐dose methotrexate followed by consolidation

whole‐brain radiotherapy and cytarabine has acceptable efﬁcacy,

great tolerability, and low toxicity in newly diagnosed PCNSL patients.

EA – previously submitted to EHA 2021.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Extranodal

non‐Hodgkin lymphoma, Combination Therapies

No conﬂicts of interests pertinent to the abstract.

323 |FOTEMUSTINE‐BASED THERAPY IN COMBINATION

WITH RITUXIMAB AS FIRST‐LINE INDUCTION CHEMOTHERAPY

FOLLOWED BY WBRT FOR NEWLY DIAGNOSED PCNSL: A

PROSPECTIVE PHASE II TRIAL

J. Wu

, F. Gao

, L. Zhang

, X. Li

, L. Li

, Z. Sun

, X. Wang

, X. Fu

, X.

Zhang

, M. Zhang

The ﬁrst afﬁliated Hospital of Zhengzhou University, Oncology

Department, Zhengzhou, China

Aim: To evaluate the safety, efﬁcacy, and feasibility of the rituximab,

fotemustine, pemetrexed and dexamethasone (R‐FPD) regimen fol-

lowed by whole‐brain radiotherapy (WBRT) for primary central

nervous system lymphoma (PCNSL) patients.

Methods: A prospective, single‐centre, phase II clinical trial was

conducted. Patients with newly diagnosed PCNSL were diagnosed at

the First Afﬁliated Hospital of Zhengzhou University from July

2018 to July 2020. The R‐FPD regimen consisted of rituximab (375

mg/m2 i.v. on D0), fotemustine (100 mg/m2 i.v. on D1), pemetrexed

(600 mg/m2 i.v. on D1), and dexamethasone (40 mg i.v. on D1‐5).

SUPPLEMENT ABSTRACTS

415

Then, WBRT was implemented based on the patient's condition

(NCT04083066).

Results: Thirty patients were included. After two cycles, the objective

response rate (ORR) was 96.4% (27/28, 26 PR, 1 CR, 0 SD, 0 PD, and 2

censored) and the disease control rate (DCR) was 96.4% (27/28). After

four cycles, the ORR was 71.4% (15/21, 5 PR, 10 CR, 1 SD, 5 PD, 7 NR,

and 2 censored) and the DCR was 76.2% (16/21). After WBRT, the

ORR was 72.8% (8/11, 3 PR, 5 CR, 1 SD, and 2 PD) and the DCR was

81.8% (9/11). The median progression‐free survival (PFS) was 20.3

months (95% CI: 5.2‐35.4). The median overall survival (OS) was 22.0

months (95% CI: 16.1‐27.9). The grade III‐IV toxicities were mainly

leukopenia (20.0%), thrombocytopenia (23.3%) and anaemia (10.0%).

Conclusion: Fotemustine‐based therapy in combination with ritux-

imab chemotherapy followed by WBRT can improve outcomes with

PFS and OS beneﬁts and has better tolerability for newly diagnosed

PCNSL patients.

The research was funded by: The research was funded by the

Natural Science Foundation of Henan Province (No. 162300410304).

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Chemo-

therapy, Combination Therapies

No conﬂicts of interests pertinent to the abstract.

324 |FRACTIONATED TOTAL‐BODY IRRADIATION AND

CYCLOPHOSPHAMIDE FOLLOWED BY AUTOLOGOUS STEM‐

CELL SUPPORT IN PATIENTS WITH T‐CELL LYMPHOBLASTIC

LYMPHOMA

M. Zhang

, R. X. Fu

, R. J. Ge

, C. Z. Sun

, F. F. Nan

The ﬁrst afﬁliated Hospital of Zhengzhou University, Oncology

Department, Zhengzhou, China

Purpose: To evaluate a treatment regimen of fractionated total body

irradiation (TBI) and cyclophosphamide (CTX) followed by autologous

stem‐cell transplantation (ASCT) in patients with T‐cell lymphoblastic

lymphoma.

Patients and Methods: 14 patients with T‐cell lymphoblastic lym-

phoma received 10.0 Gy of fractionated TBI, and CTX 60 mg/kg/day

for two days (120mg/kg total), followed by infusion of autologous

hematopoietic stem cells.

Results: All of 14 patients are alive an average of 15 (range, 1.7 to

52.3) months after transplant. The 2‐year Kaplan‐Meier (K‐M) esti-

mates of survival, progression survival (PFS) for all 14 patients were

75.2%, 58.4%, respectively. Eight (57.14%) patients had ongoing CRs

at the last disease assessment before the data cutoff. The median

time to achieve a sustained neutrophil count greater than 0.5 �10

was 12 days (range, 10 to 21).

Besides, one patients progressed after ASCT were eligible to

undergo chimeric antigen receptor T‐cell therapies.

Conclusion: Improved OS and lower relapse risk were observed

following TBI plus cyclophosphamide. We therefore recommend

TBI plus cyclophosphamide following autologous stem‐cell

transplantation for patients with T‐cell lymphoblastic lymphoma

who can't receive allogeneic HSCT.

The research was funded by: National Science and Technology

Major Project of China (Grant NO. 2020ZX09201‐009)

Keywords: Aggressive T‐cell non‐Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

325 |COMBINATION OF DECITABINE AND MODIFIED DHAP

REGIMEN: A POTENTIAL SALVAGE REGIMEN FOR RELAPSED/

REFRACTORY DIFFUSE LARGE B‐CELL LYMPHOMA AFTER

SECOND‐LINE TREATMENT FAILURE

J. Hu

, X. Wang

, M. Zhang

, Q. Chen

, X. Zhang

Department of Oncology, The First Afﬁliated Hospital of Zhengzhou

University, No.1 Jianshe East Road, Zhengzhou, Henan, China,

Department of Oncology, Zhengzhou, China

Objective: The prognosis for patients with relapsed or refractory

diffuse large B‐cell lymphoma (R/R‐DLBCL) after second‐line treat-

ment failure is extremely poor. This study prospectively observed the

efﬁcacy and safety of decitabine with a modiﬁed cisplatin, cytarabine,

and dexamethasone (DHAP) regimen in R/R‐DLBCL patients who

failed second‐line treatment.

Methods: Twenty R/R‐DLBCL patients were enrolled and treated

with decitabine and a modiﬁed DHAP regimen. The primary endpoints

were overall response rate (ORR) and safety. The secondary end-

points were progression‐free survival (PFS) and overall survival (OS).

Results: ORR reached 50% (complete response (CR) rate, 35%), ﬁve

patients (25%) had stable disease (SD) with disease control rate (DCR)

of 75%. Subgroup analysis revealed patients over ﬁfty years old had a

higher CR rate compared to younger patients (p=0.005), and relapsed

patients had a better CR rate than refractory patients (p=0.031).

Median PFS was 7 months (95% conﬁdence interval (CI), 5.1‐8.9

months). Median OS was not achieved. One‐year OS was 59.0% (95%

CI, 35.5%‐82.5%), and two‐year OS was 51.6% (95% CI, 26.9%‐76.3%).

The main adverse events (AEs) were grade 3/4 hematologic toxicities

such as neutropenia (90%), anemia (50%), and thrombocytopenia

(70%). Other main non‐hematologic AEs were grade 1/2 nausea/vom-

iting (40%) and infection (50%). No renal toxicity or death occurred.

Conclusion: Decitabine with a modiﬁed DHAP regimen can improve

the short response and prognosis of R/R‐DLBCL patients with good

tolerance to AEs, suggesting this regimen has potential as a possible

new treatment option for R/R‐DLBCL patients after second‐line

treatment failure. (identiﬁer: NCT03579082)

The research was funded by: Grant No. 81700187, 82070210

and 2020ZX09201‐009

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Chemo-

therapy, Ongoing Trials

No conﬂicts of interests pertinent to the abstract.

416

SUPPLEMENT ABSTRACTS

326 |SALVAGE RADIOTHERAPY ASSOCIATES WITH DURABLE

RESPONSE FOR A SUBSET OF PATIENTS WITH LIMITED STAGE

REFRACTORY DIFFUSE LARGE B‐CELL LYMPHOMA

J. H. Miller

, M. Gilbertson

, M. P. MacManus

, A. Wirth

, S. S. Opat

G. P. Gregory

Monash Health, Monash Haematology, Melbourne, Australia,

Peter

MacCallum Cancer Centre, Radiation Oncology, Melbourne, Australia

Introduction. Despite favourable outcomes with chemo-

immunotherapy for the majority of patients with diffuse large B‐cell

lymphoma (DLBCL), approximately one‐third of patients experience

relapsed or refractory disease portending poor prognosis. Patients

ineligible for treatment intensiﬁcation with salvage chemotherapy

may be offered palliative chemotherapy or radiotherapy, and in the

setting of single‐site progression, salvage radiotherapy with curative

intent. There remains a paucity of evidence for the latter with no

randomised clinical trials or large series reported in the literature.

The study was conducted to assess the outcomes of patients treated

with salvage radiotherapy for limited stage relapsed / refractory

DLBCL (rrDLBCL) at Monash Health, Australia, to determine its

utility and inform future treatment algorithms.

Methods. DLBCL cases (867 patients) diagnosed per WHO criteria

at Monash Health, Melbourne, Australia, between 1996 ‐2019

within the Haematology Database were analysed. Adult patients

with rrDLBCL (including transformed indolent lymphoma) treated

with salvage radiotherapy alone with curative intent were

included in the study. Patients with inadequate clinical information

or follow up data available or those in whom radiotherapy was

administered as consolidation or clearly palliative intent were

excluded.

Results. From January 1996 ‐December 2019, 18 eligible patients

were identiﬁed for inclusion in the study (12 refractory and 6

relapsed). A further 14 patients were excluded due to radiotherapy

for clearly palliative intent (n =7), CNS relapse (n =5), consolidation

(n =1) or prior indolent lymphoma (n =1). Median age at radio-

therapy commencement for the study population was 71 years (range

42‐83). 11/18 had biopsy‐proven refractory disease, a further 3/18

had inconclusive biopsy with high suspicion of active disease on im-

aging and remaining patients were treated on basis of high clinical

suspicion from PET imaging with sites not amenable to biopsy. Me-

dian radiotherapy dose was 42 Gy (range 10‐50). After a median

follow up of 14 months (range 1‐64), 11/18 patients progressed

(Fig. 1) after a median of 2 months (range 0‐20). 7/18 (39%) patients

remain in an ongoing remission without any progression events after

a median of 29 months (range 13‐64). Of the primary refractory

subset, 4/12 (33%) remain free from progression with a median

follow up of 32 months (range 13‐64).

Conclusion. Salvage radiotherapy with curative intent confers

meaningful disease control for a subset of patients with limited stage

rrDLBCL. Despite the small sample size and retrospective nature of

the study, we conclude salvage radiotherapy should be considered a

treatment option for patients ineligible for salvage chemotherapy in

cases of single site rrDLBCL. These ﬁndings warrant further clinical

investigation.

Keywords: Radiation Therapy

No conﬂicts of interests pertinent to the abstract.

327 |EFFICACY OF SALVAGE RADIOTHERAPY IN PATIENTS

WITH RESIDUAL OR RECURRENT DIFFUSE LARGE B‐CELL

LYMPHOMA

G. Cosman

, S. Dickson

, V. Chin

, S. Thompson

, S. Gupta

, Y. Chin

St George Hospital, Radiation Oncology, Sydney, Australia,

Calvary

Mater Hospital, Radiation Oncology, Newcastle, Australia,

Prince of

Wales Hospital, Radiation Oncology, Sydney, Australia

Introduction: Residual or recurrent diffuse large B‐cell lymphoma

(DLBCL) following treatment with systemic chemotherapy is usually

treated with systemic therapy ±autologous stem cell trans-

plantation. Single, or limited sites of residual or recurrent disease

may however be treated with external beam radiotherapy (EBRT).

We are presenting our results of patients with residual or recurrent

DLBCL treated with salvage EBRT alone to assess its efﬁcacy as a

salvage technique.

Methods: Patients diagnosed with DLBCL and treated with systemic

chemotherapy who had either persistent or recurrent disease were

retrospectively reviewed from databases across three tertiary hospi-

tals in New South Wales, Australia. Inclusion criteria were patients

aged ≥18 years old, diagnosis of DLBCL made between 2008‐2016,

received ﬁrst line chemotherapy, residual or recurrent disease on post‐

FIGURE 1 Event‐free survival from radiotherapy

commencement

SUPPLEMENT ABSTRACTS

417

chemotherapy ﬂuoro‐deoxyglucose positron emission tomography

(FDG PET) scan, and curative intent salvage EBRT with dose ≥30Gy.

The primary outcome was freedom from treatment failure (FFTF) post

salvage radiotherapy. Secondary outcomes included local control and

overall survival post salvage radiotherapy. Survival was calculated

from the date of radiotherapy commencement to the date of last

follow‐up.

Results: 46 patients met the inclusion criteria. Median age was 68

years (range, 19‐86 years). Just over half the patients had IPI score of

≥3 (51%) and 80% had bulky disease at initial diagnosis. The majority

of patients received 6x cycles of R‐CHOP chemotherapy (61%).

Median radiotherapy dose delivered was 40 Gy (range, 30‐46 Gy).

FFTF was achieved in 27 patients (59%) with a median follow‐up of

54 months. Local control rate and 3‐year overall survival post salvage

EBRT were 80% and 59%, respectively.

Conclusion: EBRT is an effective salvage technique for DLBCL pa-

tients with localised persistent or recurrent PET positive disease post

systemic chemotherapy, if safely encompassable by radiotherapy

ﬁelds. The challenge for the future is to determine which group of

patients have micrometastatic disease at conclusion of chemotherapy

as current functional imaging is insufﬁcient.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Radiation

Therapy

No conﬂicts of interests pertinent to the abstract.

328 |PIXANTRONE (PIX) IN RELAPSED/REFRACTORY DIFFUSE

LARGE B‐CELL LYMPHOMA (R/R DLBCL). REAL LIFE EXPERIENCE

IN 27 PATIENTS

A. Greco

, L. Tonialini

, P. R. Scalzulli

, G. De Santis

, E. Prete

, E. La

Rosa

, N. Cascavilla

, G. Tarantini

, V. Pavone

Pia Fondazione Card. G. Panico Hospital, Unit of Hematology, Tricase,

Italy,

Casa Sollievo della Sofferenza Hospital, Unit of Hematology, San

Giovanni Rotondo, Italy,

Monsignor Raffaele Dimiccoli Hospital, Unit of

Hematology, Barletta, Italy

Introduction: Treatment of relapsed/refractory (R/R) diffuse large B‐

cell lymphoma (DLBCL) represents an unmet clinical need due to the

lack of standard therapeutic options. Pix is an aza‐anthracenedione,

approved in Italy for the treatment of R/R DLBCL. Has been shown to

have anti‐tumour activity but well tolerated with signiﬁcantly

reduced cardiotoxicity. The main objective was to evaluate the ac-

tivity of this regimen in a large series of R/R DLBCL patients in a real

life setting on behalf of the Lymphoma working party of the Apulian

Hematological Network (REP).

Methods: This is a retrospective study on patients with R/R DLBCL

diagnosed from 2009 through 2020 in three hematological depart-

ment of REP. We included only patients treated with Pix as third (3L)

or fourth line (4L) from 2019 through 2020.

Results: We evaluated 27 patients (12 males and 15 females) treated

with pixantrone (13 relapsed and 14 refractory) for a median of 4

cycles (range 1–6). 59% of patients receiving treatment as third line.

The median age of patients receiving PIX was 71 years (inter‐quartile

range 43–88) and 63% had a CIRS score >6. 8 patients were diag-

nosed as evolved by low grade lymphoma, 2 as double expressor, and

1 as double‐hit DLBCL. 62% of patients had a germinal center B‐cell‐

like (GCB) subgroups. The median duration of follow up was 8.3

months. The overall response rate was 66% with complete response

rate of 33%; 40% of patients eligible for transplant used pix as bridge

to transplant. The median duration of response was 5,8 months. No

patients had grade IV adverse events, which caused drug discontin-

uation or cardiotoxicity.

Conclusion: Our data documented drug efﬁcacy that is satisfactory

for this high‐risk subset of patients with an acceptable toxicity pro-

ﬁle. Most patients who received pix were older and with elevated

CIRS‐score, which suggests the regimen may be favored also in frail

patients and transplant‐ineligible population. Pix could be one of the

most appropriate agents to bridge patients to CAR‐T‐cell therapy

based or to Transplant on its capacity to achieve rapid remission and

with safety proﬁle

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Chemotherapy

No conﬂicts of interests pertinent to the abstract.

329 |R‐GDP SCHEDULE IN PATIENTS WITH REFRACTORY OR

RELAPSED B‐CELL NON‐HODGKIN LYMPHOMA (B‐NHL)

J. Rovira

, E. González‐Barca

, J. M. Sancho

, N. Kelleher

, M.

Rodríguez

, L. Fox

, R. Parody

, S. Martin

, A. Vicent

, J. Villarroel

, C.

de la Fuente

, J. M. Ribera

, A. Sureda

, L. Escoda

Institut Català Oncologia, Hospital Joan XXIII, Universitat Rovira i Virgili,

Hematology, Tarragona, Spain,

Institut Català Oncologia, Hospital Duran

i Reynals, Universitat de Barcelona, Hematology, L'Hospitalet de

Llobregat, Barcelona, Spain,

Institut Català d'Oncologia, Hospital

Germans Trias i Pujol, Hematology, Badalona, Spain,

Institut Català

d'Oncologia, Hospital Trueta, Hematology, Girona, Spain,

Institut Català

d'Oncologia, Hospital Verge de la Cinta, Hematology, Tortosa, Spain,

Hospital Universitari Vall d'Hebron, Hematology, Barcelona, Spain

Introduction: Despite long‐term responses to ﬁrst‐line immunoche-

motherapy 30‐40% patients with B‐cell lymphomas (B‐NHL) have

relapsed/refractory (R/R) disease. Autologous stem‐cell transplant

(SCT) remains the standard of care, the optimal choice of salvage

therapy being still unknown. R‐GDP [rituximab 375 mg/m

iv, gem-

citabine 1000 mg/m

iv (2 days), cisplatin 75 mg/m

iv and dexa-

methasone 40 mg orally (4 days) every 21 days] has shown similar

effectiveness compared to other schemes with less toxicity and al-

lows adequate mobilization of stem cells to peripheral blood with the

addition of G‐CSF.

Patients and Methods: Retrospective study of efﬁcacy, toxicity and

stem‐cell mobilization capacity of R‐GDP as a salvage treatment in R/

R B‐NHL patients treated between 2018 and 2020 in 6 centers of

Catalonia.

418

SUPPLEMENT ABSTRACTS

Results: A total of 57 patients (41M:16F, median age at R‐GDP rescue

of 55 years [22‐76]) were included: 36 diffuse large B‐cell lymphoma

(DLBCL), 11 other aggressive B‐NHL and 10 follicular lymphoma; 10

DLBCL transformed from an indolent lymphoma. 81% of patients

received R‐GDP as 2

line. Median time from diagnoses to R‐GDP

administration was 10 months (2‐175) and most of the patients

(56%) received 3 cycles. Grade 3‐4 toxicity was observed in 42% of

cases, mostly hematological (table). Hospital admission was required in

18% of cases with a median stay of 7 days. Main reason for admission

was infectious complications (42% neutropenic fever) despite pro-

phylactic G‐CSF in 85% of cases. Response was: complete remission

(CR) 18 (32%), partial remission 11 (19%), progression 28 (49%).

Mobilization with G‐CSF (10 µg/Kg/12h, median 5 days) was carried

out after 3

R‐GDP cycle. Most patients (n =20, 80%) required 1 day of

apheresis, 26% (n =7) plerixafor use and the median CD34/Kg was 4

x10

(2,2‐15,3). Only 2 patients (3,6%) showed inadequate mobiliza-

tion. Only 21 patients of 55 candidates (38%) reached the auto‐SCT

procedure, 6 patients despite R‐GDP refractoriness (69% in CR).

70% non‐SCT patients were refractory. Median days of neutrophil and

platelet engraftment were 10 (7‐18) and 11 (8‐14), respectively. Four

patients proceeded to allo‐SCT and 10 received CAR‐T cell therapy (8

no SCT and 2 relapses post auto‐SCT). 2‐year progression free and

overall survival from R‐GDP were 22 and 36%, respectively.

Conclusion: R‐GDP can be used as a successful mobilization (plus G‐

CSF) and well‐tolerated salvage regimen for R/R B‐NHL patients

candidates to auto‐SCT with similar tumor reduction efﬁcacy than

other schemes.

Table: Patient treatment characteristics and R‐GPD toxicity

(n =57)

Previous R‐GDP treatment (%)

1 81

2 10

≥3 9

R‐GDP cycles (%)

1 14

2 25

3 56

4 5

R‐GDP toxicity (%)

Grade 1‐2 81

Gastrointestinal 41

Fatigue 38

Thrombosis 6

Hematological 13

Anemia 6

Neutropenia 6

Thrombocytopenia 8

(Continues)

(Continued)

Previous R‐GDP treatment (%)

1 81

2 10

≥3 9

Grade 3‐4 42

Gastrointestinal 2

Fatigue 5

Neurological 2

Hematological 91

Anemia 14

Neutropenia 18

Thrombocytopenia 50

Figure: Progression free survival (PFS) and Overall survival

(OS) from R‐GDP salvage treatment

EA – previously submitted to regional or national meetings (up to

1000 attendees).

Keywords: Chemotherapy, Stem Cell Transplant

No conﬂicts of interests pertinent to the abstract.

330 |POLATUZUMAB VEDOTIN PLUS BENDAMUSTINE AND

RITUXIMAB IN PATIENTS WITH RELAPSED/REFRACTORY

DIFFUSE LARGE B‐CELL LYMPHOMA IN THE REAL WORLD

P. Vodicka

, K. Benesova

, A. Janikova

, V. Prochazka

, D. Belada

, H.

Mocikova

, K. Steinerova

, J. Duras

, J. Karban

, V. Hanackova

, A.

Sykorova

, A. Obr

, M. Trneny

Charles University and General University Hospital, First Department of

Medicine, Prague, Czech Republic,

Faculty of Medicine, Masaryk

University and University Hospital, Department of Haematology and

Oncology, Brno, Czech Republic,

Faculty of Medicine and Dentistry,

Palacky University and University Hospital, Department of Haemato‐

SUPPLEMENT ABSTRACTS

419

Oncology, Olomouc, Czech Republic,

Faculty of Medicine, Charles

University and University Hospital, 4th Department of Internal Medicine ‐

Haematology, Hradec Kralove, Czech Republic,

Third Faculty of Medi-

cine, Charles University and University Hospital Kralovske Vinohrady,

Department of Internal Medicine – Haematology, Prague, Czech Republic,

Faculty of Medicine, Charles University and University Hospital,

Department of Haemato‐Oncology, Pilsen, Czech Republic,

Medical

Faculty of the Ostrava University and University Hospital, Department of

Haematology, Ostrava, Czech Republic

Introduction. The treatment combination of polatuzumab vedotin,

bendamustine and rituximab (Pola‐BR) has been recently approved

for transplant‐ineligible patients with relapsed/refractory diffuse

large B‐cell lymphoma (R/R DLBCL). However, the number of pa-

tients treated in the GO29365 trial (NCT02257567) including the

extension cohort was limited, and more data is needed. We retro-

spectively analysed the outcome of 21 patients treated with pola-

tuzumab vedotin to determine the efﬁcacy and safety of this

treatment regimen in the real‐world settings.

Methods. A total of 21 patients were scheduled to receive 6 cycles

of Pola‐BR every 21 days (polatuzumab vedotin 1.8 mg/kg; ritux-

imab 375 mg/m

; bendamustine 70 mg/m

in 7 cases and 90 mg/m

in 13 cases; no bendamustine in 1 case). Data of all patients were

entered into the database of the NiHiL project (NCT03199066).

Treatment‐related adverse events (AEs) were classiﬁed according to

the Common Terminology Criteria for Adverse Events. Primary

endpoint of the study was the overall survival (OS), secondary

endpoints included the progression‐free survival (PFS), and

response rate to the therapy.

Results. Median age at relapse was 67 years (range 35‐85 years).

Sixteen (76.2%) patients were diagnosed with clinical stage III‐IV, 10

(47.6%) patients with performance status according to ECOG ≥2, 16

(76.2%) patients with elevated levels of lactate dehydrogenase and

with extranodal involvement, 15 (71.4%) patients with high‐

intermediate or high‐risk International Prognostic Index. Patients

were pre‐treated with median of 3 previous lines of therapy (range

2‐7) including autologous stem cell transplant in 6 (28.6%) cases.

Median number of Pola‐BR cycles was 2 (range 1‐6), 5 (23.8%) pa-

tients received all intended 6 cycles of the therapy. Median OS from

relapse was 8.7 months, median PFS was 3.8 months, with a median

follow‐up of living patients of 6.8 months. The overall response rate

was 33.3% (n =7), 5 (23.8%) patients reached complete remission, 2

(9.5%) patients reached partial remission. Stable disease was

detected in 2 (9.5%) cases, progressive disease (PD) in 9 cases

(42.9%). Grade 3‐4 haematologic toxicities included neutropenia in 6

(28.6%) patients, thrombocytopenia in 8 (38.1%) cases, and anaemia

in 4 (19.0%) cases. Peripheral neuropathy was detected in one (4.8%)

patient, infections grade 3‐4 in 5 (23.8%) cases. In 16 (76.2%) pa-

tients, the reason for early discontinuation of the treatment included

PD (n =8; 50.0%), followed by AEs (n =5; 31.3%), and intended

bridging to CAR‐T therapy (n =3; 18.8%).

Conclusions. The use of Pola‐BR regimen in the real world demon-

strates tolerable toxicity proﬁle and efﬁcacy in transplant‐ineligible

patients with R/R DLBCL. Moreover, this regimen might represent

a perspective option as a bridge to CAR‐T therapy.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Chemo-

therapy, Immunotherapy

Conﬂicts of interests pertinent to the abstract

p. vodicka

Consultant or advisory role: Roche

A. Janikova

Consultant or advisory role: Roche

D. Belada

Consultant or advisory role: Roche

M. Trneny

Consultant or advisory role: Roche

331 |TIME‐TO‐RESPONSE FOR PATIENTS WITH RELAPSED/

REFRACTORY AGGRESSIVE B CELL NON‐HODGKIN LYMPHOMA

TREATED WITH POLATUZUMAB‐BASED THERAPY

M. E. Hughes

, S. D. Nasta

, J. N. Gerson

, J. Svoboda

, E. A. Chong

S. J. Schuster

, S. K. Barta

, K. W. Robinson

, D. J. Landsburg

University of Pennsylvania, Abramson Cancer Center, Philadelphia,

Pennsylvania, USA

Introduction: Knowledge of time‐to‐response after initiation of ther-

apy (TTR) may inﬂuence treatment recommendations made to patients

(pts) with relapsed/refractory (R/R) aggressive B cell non‐Hodgkin

lymphoma (aBNHL) given the potential for rapid disease growth. If

reported in clinical trials, TTR is dependent upon the timing of

protocol‐speciﬁed imaging response assessments (IRA) and does not

account for clinical response to therapy (Rc), which may precede

radiographic response (Rr). Here we report TTR and other response‐

related outcomes in pts with R/R aBNHL treated with polatuzumab

(pola).

420

SUPPLEMENT ABSTRACTS

Methods: Pts analyzed were those with non‐Burkitt R/R aBNHL

diagnosed on most recent tissue biopsy and ﬁrst treated with pola

at the University of Pennsylvania from 7/2019 to 12/2020. Rc was

deﬁned as reduction in lymph node size on examination, resolution

of disease‐related symptoms, normalization of lactate dehydroge-

nase (LDH) previously >2x upper limit of normal or disease

response on imaging performed for indication other than IRA. Rr

was assessed per Revised Response Criteria for Malignant Lym-

phoma. Timing of clinical/laboratory assessments and IRA were at

the discretion of the treating clinician. Data were censored on 3/1/

2021.

Results: Fifty pts were included in this analysis. Baseline clinico-

pathologic and treatment characteristics at time of pola initiation

are listed in Table 1. Nineteen pts (38%) were treated with the

intent to bridge to cellular therapies (CT). Median number of cycles

of pola received was 3. Rc was achieved by 18 pts (36%), Rr by 16

pts (32%) and the composite outcome of Rc or Rr (overall response,

[Ro]) by 24 pts (48%). IRA was not performed for 9 pts: 7 due to

proceeding to planned CT (5 with Rc) and 2 due to death not

attributed to aBNHL with ongoing Rc. The median TTRc, TTRr and

earliest of TTRc or TTRr (TTRe) were 21 d (range 7‐41 d), 64

d (range 17‐137 d) and 22 d (7‐67 d), respectively. Of 12 pts who

achieved Rc and had subsequent IRA performed, 8 achieved Rr.

Twelve pts stopped tx in remission: 7 pts due to proceeding to

planned CT, 3 due to toxicity (gastrointestinal in 2 pts and neu-

ropathy in 1 pt) and 2 pts due to choice. Excluding the 7 pts

proceeding to planned CT in remission, the estimated proportion of

pts with ongoing Rc and Rr at 180 d were 53% and 51%, respec-

tively. Of characteristics listed in Table 1, non‐high grade B cell

histology and receipt of prior CD19‐directed chimeric antigen

receptor‐modiﬁed T cell therapy were signiﬁcantly associated with

achievement of Ro; additionally, age ≤60 years, largest tumor

diameter ≤7.5 cm and disease refractory to most recent prior

therapy were signiﬁcantly associated with shorter TTRe.

Conclusions: For R/R aBHL pts treated with pola at our institution,

frequently with the intention to bridge to CT, 48% achieved Rc or Rr

with median TTR of 22 d. These ﬁndings may further inform use of

pola in the standard‐of‐care and investigational settings.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma

Conﬂicts of interests pertinent to the abstract

M. E Hughes

Consultant or advisory role: AstraZeneca, Genzyme, Janssen, AbbVie,

Karyopharm

Research funding: Acerta

S. D Nasta

Consultant or advisory role: Morphosys, Merck

Research funding: Roche/Genentech, Millennium/Takeda, Pharma-

cyclics, ATARA, Forty Seven

J. N Gerson

TABLE 1Baseline characteristics

SUPPLEMENT ABSTRACTS

421

Consultant or advisory role: Pharmacyclics, Genentech, AbbVie

Research funding: Loxo

J. Svoboda

Consultant or advisory role: Seattle Genetics, Bristol‐Myers Squibb,

Pharmacyclics, Imbrium Therapeutics, Genmab, Adaptive Bio-

technologies, ADC Therapeutics, Atara Biotherapeutics

Research funding: Celgene, Seattle Genetics, Pharmacyclics, Merck,

Bristol‐Myers Squibb, Incyte

Educational grants: Imbrium Therapeutics

E. A Chong

Consultant or advisory role: Novartis, Tessa Therapeutics, Bristol‐

Myers Squibb, Kite/Gilead

S. J Schuster

Consultant or advisory role: Celgene, Nordic Nanovector, Novartis,

AbbVie, Acerta, Alimera Sciences, BeiGene, Juno Therapeutics, Loxo

Oncology, Tessa Therapeutics, Genetech/Roche, Research funding:

Novartis, Pharmacyclics, Adaptive Biotechnologies, Merck, Gene-

tech/Roche, Celgene, Juno Therapeutics, AbbVie, Incyte, TG Thera-

peutics, DTRM,

S. K Barta

Consultant or advisory role: Monsanto

Honoraria: Atara, Seattle Genetics, Janssen, Pﬁzer, Acrotech

Research funding: Seattle Genetics,

D. J Landsburg

Consultant or advisory role: Morphosys, Karyoparhm, Celgene

Research funding: Takeda, Curis, Triphase

Other remuneration: Karyopharm

332 |PET ORIENTED INTENSIFICATION AND MAINTENANCE

WITH RITUXIMAB IN FIRST LINE NON‐HODGKIN LARGE B CELL

LYMPHOMA (DLBCL)

L. Pezzullo

, G. Cassiordor0

, R. Rosamilio

, I. Ferrara

, S. Luponio

, B.

Serio

, C. Martorelli

, L. Mettivier

, R. Fontana

, R. Guariglia

, C.

Selleri

Salerno University, Hematology Division, Salerno, Italy

The addition of rituximab to induction chemotherapy improved the

outcome of patients with diffuse large B‐cell lymphoma (DLBCL).

However, relapse prevention with anti‐CD20 antibody maintenance

therapy is not the standard in DLBCL lymphoma conﬁrmed by

various randomized trials.

To study in patients who achieved an uncertain or unsatisfactory

response (PR or CRu guided by PET with a deauville score of 2‐3

after induction therapy with R‐CHOP or similar chemotherpy) with

intensiﬁcation therapy with rituximab (Rituximab 375 mg / m2 per

week for 4 weeks) followed by maintenance (rituximab 375 mg / m2

every 2 months for a total of 12 administrations or until any

progression or unacceptable toxicity) to assess whether we observe

an improvement in results in this cohort of patients.

From January 2014 to January 2021 we studied 75 consecutive

patients with DLBCL (45m and 30f, mean age: 65 years (range 28‐82

years); 37 ABC, 25 GC; 5 NOS; 5 T‐rich and 3 immunoblastic), treated

with R‐CHOP chemotherapy cycles or similar.

Of the 75 patients (with a median follow‐up of 30 months) the ORR

was 77% (58/75 patients: CR 39; CRu 16 and PR 3); 7 NR (9%); 7 (9%)

died before completion of therapy and 3 (5%) has still on R‐CHOP

treatment. The total OS, PFS and EFS of this group projected at 54

months was respectively of 75%; 78% and 75%.

After induction therapy with R‐CHOP 58/75 patients (77%)

responded to treatment (39 CR; 16 CRu and 3 PR) and we considered

these patients to evaluate responses to intensiﬁcation and mainte-

nance treatment with rituximab. oriented by the PET response at the

end of R‐CHOP.

Results: we treated 18/58 patients 31% of patients, 16 in CRu (ﬁnal

PET score 2‐3 deauville) and 2 in PR after R‐CHOP induction with

consolidation and maintenance with rituximab, 13 m and 5 f with

median age age 66 (range 44‐77) 8 ABC; 7 GC; 2 immunoblastic and

1 rich in T. No signiﬁcant side effects were observed in this patient

group.

After a median follow‐up of 40 months (range 8‐88 months) 17/18

(94%) patients are in CR (only one patient relapsed after 18 months).

While after a median follow‐up of 30 months of 40/58 patients in CR

after induction of R‐CHOP, 33 (83%) are in CR while 7 (17%) had a

relapse.

Observations of these 2 groups projected at 85 months were similar

for OS (94% and 93% respectively); while PFS and EFS were 94% and

82% in the maintenance or non‐maintenance groups, respectively.

Summary / Conclusion: these retrospective data, collected by our

center, demonstrate an improvement in PFS and EFS in patients with

RP or CRu with DLBCL compared to the literature in particular, an

overall advantage was noted for the DLBCL ABC subgroup per-

forming this treatment ( total patients DLBCL ABC and GC with OS

at 70 months 82% and 76% respectively).

This improvement may be due to consolidation and rituximab

maintenance therapy in the deuville 2‐3 score cohort after R‐

CHOP induction (PET‐oriented treatment). A larger cohort of pa-

tients and a randomized trial are needed to conﬁrm these pre-

liminary data

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

333 |INTRAVENOUS HIGH DOSE METHOTREXATE AS

CENTRAL NERVOUS SYSTEM (CNS) PROPHYLAXIS IN PATIENTS

WITH DIFFUSE LARGE B CELL LYMPHOMA (DLBCL) WITH HIGH

RISK FOR CNS PROGRESSION

J. A. Villarreal

, J. Rovira

, M. Franch

, M. Encuentra

, D. Blazevic

M. Rodriguez

, N. Kelleger

, S. Martín

, A. C. Oliveira

, E. Domingo

422

SUPPLEMENT ABSTRACTS

Domenech

, J. M. Ribera

, A. Sureda

, J. M. Sancho

, L. Escoda

, E.

González‐Barca

Institut Català d'Oncologia, Hospital Duran i Reynals, Institut

d'Investigació Biomèdica de Bellvitge (IDIBELL), Hematology,

L'Hospitalet de Llobregat, Barcelona, Spain,

Institut Català d'Oncologia,

Hospital Universitari Joan XXIII, Hematology, Tarragona, Spain,

Institut Català d'Oncologia, Hospital Germans Trias i Pujol, Institut de

Recerca contra la Leucèmia Josep Carreras, Universitat Autònoma de

Barcelona, Hematology, Badalona, Barcelona, Spain,

Institut Català

d'Oncologia, Hospital de Tortosa Verge de la Cinta, Hematology,

Tortosa, Tarragona, Spain,

Institut Català d'Oncologia, Hospital

Universitari de Girona Doctor Josep Truet, Institut d'Investigació

Biomèdica de Girona (IDIBGI), Universitat de Girona, Hematology,

Girona, Spain

Introduction: Despite improvement in survival in diffuse large B‐cell

lymphoma (DLBCL), central nervous system (CNS) relapse represents

a clinical challenge. The most widely used CNS prophylaxis is intra-

thecal methotrexate, which arrives to the leptomeninges, but it has

limited value in DLBCL, which usually presents CNS parenchymal

involvement. Therefore, prophylaxis should include deeply penetrant

drugs such as intravenous high‐dose methotrexate (HDMTX). Our

aim was to evaluate the feasibility and efﬁcacy of HDMTX in com-

bination with immunochemotherapy in patients with high risk criteria

for CNS progression.

Patients and Methods: Retrospective study of 47 patients, diag-

nosed from October 2016 to July 2020, in 5 hospitals of the Catalan

Institute of Oncology in Spain, who received CNS prophylaxis with

HDMTX (3 g/m

) and who were diagnosed of: DLBCL with high risk

criteria for CNS progression [CNS International Prognostic In-

dex (CNS‐IPI) ≥4, speciﬁc extranodal sites involvement: pericardial,

bone marrow, testes, uterine, breast]; or high grade lymphoma

double hit.

Results: Characteristics at diagnosis are summarized in the table.

Of note, median age: 53 years, CNS‐IPI ≥4: 42 (89%), high risk

extranodal sites involvement: 25 (53%), double hit: 7 (15%).

Treatment consisted of 3 cycles HDMTX alternating with 6 cycles

R‐CHOP (cycles 2,4,6): 35 patients; 2 cycles HDMTX alternating

with 6 cycles R‐CHOP (cycles 3,6): 4 patients; 6 cycles R‐CHOP

followed by 3 cycles HDMTX: 2 patients. Six patients with dou-

ble hit lymphoma were treated with 3 cycles HDMTX alternating

with 6 cycles DA‐R‐EPOCH (cycles 2,4,6): 3 patients; 6 cycles DA‐

R‐EPOCH followed by 2 cycles HDMTX: 3 patients. Seven (15%)

patients presented acute kidney injury after the ﬁrst dose of

HDMTX, 3 (6.4%) grade 3‐4, in whom HDMTX was stopped (in 1

patient R‐CHOP was delayed 14 days). All patients with DA‐R‐

EPOCH completed the planned doses. Other frequent toxicities

among 112 cycles of HDMTX were: neutropenia 8 (7%), anemia 6

(5.3%), neutropenic fever 4 (3.6%), and mucositis 3 (2.7%), mostly

grade 1‐2. Eight (17%) patients stopped treatment before

completing their plan, 4 due to progression (one in CNS), 3 due to

infectious complications and 1 for unknown reason. Response was

evaluable in 45 patients: CR 34 (76%), PR 5 (11%), progression 6

(13%). With a median follow‐up of 22 months, PFS was 62.7%

(95%CI 48‐77.4) and OS was 66.4% (95%CI 49.5‐83.3). Three

(6.4%) patients progressed in CNS, two at 6 months and one at 10

months from diagnosis.

Conclusions: HDMTX is used as CNS prophylaxis in a young subgroup

of DLBCL patients. HDMTX alternating with systemic immunoche-

motherapy is feasible and can be administered safely in most cases,

being acute kidney injury the most frequent toxicity leading to

discontinuation. Nevertheless, 6% of patients had an early CNS

relapse. Therefore, new strategies should be investigated in this very

high risk population.

SUPPLEMENT ABSTRACTS

423

Figure: Progression‐free survival of 47 patients with high

risk of CNS progression treated with HDMTX

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Chemo-

therapy, Combination Therapies

No conﬂicts of interests pertinent to the abstract.

334 |SAFETY OF SAME DAY HD‐MTX WITH INDUCTION

THERAPY FOR DLBCL WITH CONCURRENT CNS DISEASE OR AS

PROPHYLAXIS FOR HIGH RISK OF CNS RELAPSE

B. Barlow

, C. D. Bodine

, H. Hatic

, G. Goyal

, A. Mehta

M. Narkhede

University of Alabama at Birmingham, Hematology/Oncology,

Birmingham, Alabama, USA

Background: High‐dose methotrexate (HD‐MTX) at a dose between

2.5 to 5 gm/m

is commonly administered in conjunction with stan-

dard induction chemotherapy to patients with Diffuse Large B Cell

Lymphoma (DLBCL) at high risk of central nervous system (CNS)

relapse, as deﬁned by the Lymphoma International Prognostic Index

(CNS‐IPI). Optimal timing of HD‐MTX in relation to induction

chemotherapy is unknown. A recent study suggested that HD‐MTX

intercalated with R‐CHOP cycles was associated with increased

toxicity and treatment delays without improvement in survival or

CNS relapse compared with end of treatment MTX (Wilson et al

2020). This retrospective study evaluates the toxicities and treat-

ment delays associated with HD‐MTX administered on day 1 of cy-

cles of chemo‐immunotherapy.

TABLE 1Outcome of Day 1 HD‐MTX with induction Therapy (n=45)

424

SUPPLEMENT ABSTRACTS

Methods: This single center retrospective cohort study included 45

patients(pts) with DLBCL with concurrent CNS disease or at high risk

of CNS relapse who received HD MTX on day 1 of chemo‐

immunotherapy at our center. Data was abstracted from chart re-

view and included variables describing clinical and treatment char-

acteristics, time to MTX clearance, toxicities experienced and

treatment delays.

Results: 31 pts received HD‐MTX on the day of R‐CHOP chemo-

therapy, 6 pts received HD‐MTX on the day of R‐EPOCH (Rituximab,

Etoposide, Doxorubicin, Vincristine, Cyclophosphamide, and Predni-

sone) and 8 pts received HD‐MTX with R‐MiniCHOP (dose reduced

R‐CHOP). Same day HD MTX with chemo‐immunotherapy was

associated with acute kidney injury (AKI; 17‐25%), treatment delays

>7 days (13‐17%), and grade 2 mucositis (11‐50%). The burden of

toxicities was numerically higher in patients treated with R‐EPOCH

vs. R‐CHOP (Table 1). Clinical outcomes are summarized in Table 1

below.

Conclusion: In our heterogeneous population of pts, we describe that

the incidence of toxicities and treatment delays experienced with

same day HD‐MTX are higher with R‐EPOCH than with R‐CHOP.

Comparative studies with intercalated or end of treatment MTX will

determine if the incidence of treatment delays, toxicities and de‐

escalations are higher with same day HD‐MTX administration.

EA – previously submitted to ASCO 2021.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

335 |R‐CODOX‐M/R‐IVAC IN HIGH RISK AND MYC MUTATED

DIFFUSE LARGE B CELL LYMPHOMA‐A SINGLE CENTRE

RETROSPECTIVE STUDY

S. L. Beverstock

, F. M. Scott

Western General Hospital, Department of Haematology, Edinburgh, UK

Introduction: Rituximab, cyclophosphamide, doxorubicin, vincristine,

prednisolone (R‐CHOP), treatment for diffuse large B cell lymphoma

(DLBCL), fails to achieve long term disease control in 50% patients

with a high (3‐5) international prognostic index (IPI). Outcome post

R‐CHOP of double (DH) or triple hit (TH) disease, MYC rearrange-

ment concurrent with BCL2, BCL6 rearrangements or both, is also

poor with 5 year progression free survival (PFS) 27% and overall

survival (OS) 18%.

UK National Cancer Research Institute (NCRI) Phase 2 Study of

efﬁcacy of 1st line rituximab, cyclophosphamide, vincristine, doxo-

rubicin, methotrexate, ifosfamide, etoposide, cytarabine (R‐CODOX‐

M/R‐IVAC) in high risk DLBCL (IPI 3‐5) achieved 2 year PFS of

67.9%. Fluorescence in situ hybridisation (FISH) for MYC, BLC2,

BCL6 was done in 51.4% of patients so ﬁndings cannot necessarily be

extrapolated to high grade B lymphoma as per 2016 World Health

Organisation criteria.

Methods: We undertook a retrospective review of patients within

South East Scotland who received R‐CODOX‐M/R‐IVAC as initial

therapy for DLCBL with high risk features (IPI 3‐5 or recurrent ge-

netic rearrangements by FISH) between 2014 and 2020. 35 patients

were identiﬁed. FISH analysis of biopsy was available in 91%: 51%

had a normal MYC, 11% had an isolated MYC translocation, 6% TH,

23% DH (17% MYC/BCL2; 6% MYC/BCL6. Immunoglobulin partner

gene status was unknown.

Results: Median age 59 years (range 24–73), 57% were male, 37%

were treated on over 65s protocol. 14 (40%) failed to complete

treatment due to: cytopenias (29%), neutropenic sepsis (21%), hep-

atoxicity (14%), progressive disease (PD) (8%), mucositis (8%).

Treatment delays occurred in 22 (63%), median duration of delay 16

days (range 3‐57): infection accounted for 74%, delayed count re-

covery 13%, hepatotoxicity 9%. There were 2 treatment related

deaths.

Outcomes were similar to those observed in NCRI study: 24

(69%) achieved complete response (CR), 1 year OS 71%. 8 (23%) had

primary refractory disease (PRD), 7 (88%) of whom died of PRD. 3

(11%) responding patients relapsed: 1 achieved CR post salvage, 2

died of PD.

Impact of MYC status was variable: CR rate was 72% with

normal MYC, 79% if isolated MYC mutation, 75% in DH and 50% in

TH. Risk of PRD was higher if MYC abnormality detected: 17% with

normal MYC relapsed compared to 50% with a MYC abnormality.

Conclusion: R‐CODOX‐M/R‐IVAC proved effective in DH and TH

disease achieving OS of 77% at 2 year, beneﬁt in other high risk

patients was less clear. Treatment‐related toxicity in the real

world setting was signiﬁcant; 40% of patients failed to complete

all 4 cycles and treatment delays were common. The local man-

agement guideline has subsequently been modiﬁed, de‐escalating

treatment to R‐CHOP for those without DH/TH or CNS

involvement.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Pathology and

Classiﬁcation of Lymphomas, Chemotherapy

No conﬂicts of interests pertinent to the abstract.

336 |ZANUBRUTINIB, LENALIDOMIDE PLUS R‐CHOP (ZR

‐

CHOP) AS THE TREATMENT FOR DIFFUSED LARGE B‐CELL

LYMPHOMA (DLBCL)

H. Zhu

, Y. Sha

, W. Wu

, R. Chen

, Y. Yang

, J. Qiu

, H. Mi

, C. Peng

C. Ding

, Z. Wang

, L. Fan

, W. Xu

, J. Li

The First Afﬁliated Hospital of Nanjing Medical University, Department

of Hematology, Nanjing, China,

Pukou division of Jiangsu Province

Hospital, Pukou CLL Center, Nanjing, China,

The First Afﬁliated Hospital

of Nanjing Medical University, Department of Nuclear Medicine, Nanjing,

China,

The First Afﬁliated Hospital of Nanjing Medical University,

Department of Pathology, Nanjing, China

SUPPLEMENT ABSTRACTS

425

Introduction: To evaluate the safety and efﬁcacy of zanubrutinib,

lenalidomide plus R‐CHOP (ZR

‐CHOP) as the treatment for DLBCL

patients, we conducted this single‐arm retrospective observational

study.

Methods: We enrolled patients aged 18 to 75 years old with high‐

risk DLBCL. Oral zanubrutinib was given continuously (160mg

twice daily) from Day 0, lenalidomide 25mg daily Day 1‐7. Patients

were administered intravenously rituximab (375mg/m

Day 0),

cyclophosphamide (750mg/m

Day 1), doxorubicin (50mg/m

Day

1), vincristine (1.4mg/m

Day 1), and oral prednisone (50mg/day

Day 1‐5). All patients were recommended to receive 6 cycles of

‐CHOP (R/R

‐CHOP were allowed in cycle 1‐2 due to poor

physical condition at treatment) and patients older than 70 years

old were administered ZR

‐miniCHOP (Figure 1). ctDNA was

dynamically detected before treatment, after 3 and 6 cycles to

evaluate tumor mutational burden. The primary endpoint was

complete response ratio (CRR) after mid‐term and 6 cycles. The

secondary endpoint was overall response rate (ORR), ctDNA and

the number of adverse events (AE). AEs were graded based on

CTCAE (version 5.0).

Results: 10 DLBCL patients diagnosed in Pukou CLL Center were

enrolled in this cohort between July 2020 and February 2021, with 9

treatment‐naïve DLBCL and 1 Relapsed/Refractory DLBCL (PD after

4 cycles of bendamustine plus rituximab). The median age of patients

was 55 years old and all patients had ECOG‐PS ≤2. 1 patient (1/10)

was diagnosed as double‐hit DLBCL and 7 patients (7/10) as double‐

expression. 8 patients were non‐GCB and 2 were GCB. 7 patients (7/

10) were classiﬁed as high‐intermediate and high‐risk group ac-

cording to NCCN‐IPI (Table 1). At data cutoff (1st March, 2021), the

median follow‐up was ﬁve months (3‐8 months) with all patients have

completed at least 3 cycles and mid‐term assessment could be con-

ducted. The ORR was 100.0%, with 9 patients achieved CR (9/10) and

1 patients achieved PR (1/10). ctDNA was dynamically detected in

ﬁve patients. The median number of baseline somatic mutation was 5

and all 5 patients showed undetectable ctDNA after 3 cycles. 4 pa-

tients have received 6 cycles, all of them achieved CR (4/4) and un-

detectable ctDNA (4/4) (Figure 1). The most common hematological

toxicity events were lymphocytes count decreased, neutrophil count

decreased, thrombocytopenia and anemia, with 3‐4 level occurrence

rate was 70.0%, 30.0%, 20.0% and 20.0%. The most common non‐

hematological toxicity events were nausea and fatigue. One pa-

tients discontinued oral zanubrutinib and lenalidomide more than

seven days due to drug rash.

Conclusion: ZR

‐CHOP for high‐risk DLBCL patients with fair

physical condition could achieve high CRR and high proportion of

early‐stage undetectable ctDNA. The overall tolerability was under

control. ZR

‐CHOP could be one of the promising choices for the

treatment of high‐risk DLBCL.

EA – previously submitted to regional or national meetings (up to

1000 attendees) and EHA 2021.

The research was funded by: This study was supported by Fund

of National Natural Science Foundation of China (Grant No.

81970146), National Science Foundation of China International

Cooperation and Exchange Programme (Grant No. 81720108002),

National Science and Technology Major Project (Grant No.

2018ZX09734007) and Six Talent Peaks Project in Jiangsu Province,

2019 (Grant No. WSN‐001).

Keywords: Chemotherapy, Molecular Targeted Therapies, Combina-

tion Therapies

No conﬂicts of interests pertinent to the abstract.

TABLE 1Baseline Clinical Characteristics of DLBCL patients

426

SUPPLEMENT ABSTRACTS

337 |PROGNOSTIC IMPACT OF NUTRITIONAL STATUS ON

DLBCL PATIENTS

C. Barros Lima

, B. Marques

, S. Duarte

, C. Afonso

, D. Neves

, A.

Lai

, M. Julião

, L. Ruzickova

, J. Carda

, M. Gomes

, A. Cipriano

, A.

Espadana

Centro Hospitalar e Universitário de Coimbra, Hematology Department,

Coimbra, Portugal,

Centro Hospitalar e Universitário de Coimbra,

Pathology Department, Coimbra, Portugal

Introduction: The prognostic impact of body mass index (BMI) in

diffuse large B cell Lymphoma (DLBCL) patients is controversial, as

some studies showed a survival beneﬁt for higher BMI while others

showed no correlation or even a detrimental effect on survival. On

the other hand, studies regarding nutritional status impact on pa-

tients outcome have shown that lower Prognostic Nutritional Index

(PNI) scores are related to lower Progression Free Survival (PFS) and

Overall Survival (OS).

Aim: Evaluate the impact of BMI and PNI at diagnosis on OS and PFS

on a cohort of patients with DLBCL, NOS.

Methods: Retrospective data from patients (pts) with DLBCL,

NOS, diagnosed between 2010 ‐2019. BMI was calculated as

weight/height2. PNI was calculated as 10xserum albumin (g/dL)

+0.005xabsolute lymphocyte count. Two groups were deﬁned

for both variables for statistical analysis using a ROC curve to

discriminate outcomes (OS and PFS). A p value less than

0.05 was considered statistically signiﬁcant. Univariate and multi-

variate analysis was performed to evaluate the impact on the

outcomes.

Results: A total of 335 patients were diagnosed with DLBCL, NOS,

with a median age of 65 yrs; 177 (53%) were male and 248 (74%) had

an ECOG‐PS 0‐1. The BMI was <18.5 Kg/m2 on 8 pts (2.4%), 18.5‐

24.9 Kg/m2 on 135 (40.3%), 25 ‐29.9 Kg/m2 on 82 (24.5%) and ≥30

Kg/ m2 on 43 (12.8%).

Using ROC curves, the cut‐offs for BMI and PNI were 31 Kg/m2 and

38, respectively. Patients with higher BMI (≥31 Kg/m2) showed

better ECOG‐PS (p =0.03) and a lower Ann Arbor stage (p =0.036).

After 5 years of follow up, patients with lower BMI (<31 Kg/ m2) had

a non statistically difference in both PFS (52% vs 40%, p =0,49) and

OS (58% vs 50%, p =0,75) vs higher BMI.

PNI was signiﬁcantly higher on male pts (p =0.02), patients under 65

years old (p =0.004), Ann‐Arbor stage I‐II (p =0.001), ECOG‐PS 0‐1

(p <0.001) and BMI ≥31 Kg/m2 (p =0.022). At 5 years, patients with

higher PNI (>38) showed a higher PFS (63% vs 32%, p <0,001) and

OS (68,5% vs 37,5%, p <0,001), both statistically relevant. However,

when taking into account other variables, namely age, ECOG‐PS, IPI,

Ann‐Arbor stage, B symptoms and low albumin, PNI impact on PFS

(p =0.065) and OS (p =0.072) was not statistically signiﬁcant,

Conclusion: In our cohort, BMI and PNI did not have a signiﬁcant

impact in PFS or OS although the impact of PNI on PFS remained close

to being signiﬁcant. More studies are needed to evaluate the impact of

nutritional status and its relevance for inclusion of PNI as a tool for

clinical evaluation and risk assessment in patients with DLBCL, NOS.

Keywords: Other

No conﬂicts of interests pertinent to the abstract.

338 |PROGNOSTIC VALUE OF RED CELL DISTRIBUTION

WIDTH AT DIAGNOSIS (RDW) IN DIFFUSE LARGE B‐CELL

LYMPHOMA.

J.M. Raya

, P. López‐García

, C. D. Reyes

, M.J. Rodríguez‐Salazar

C. De Bonis

, B. González‐González

, M. T. Hernández‐García

S. Lakhwani

Hospital Universitario de Canarias, Hematology, La Laguna, Spain,

Universidad de La Laguna, Facultad de Ciencias de la Salud, sección

Medicina, La Laguna, Spain

FIGURE 1 Treatment Scheme of ten DLBCL patients.

SUPPLEMENT ABSTRACTS

427

Diffuse large B‐cell lymphoma (DLBCL) constitutes approximately

30% of adult non‐Hodgkin lymphomas in developed countries, and

even a higher percentage in developing countries. R‐CHOP is the

standard chemotherapy regimen and 5‐year overall survival ranges

from 60 to 70%. Main prognosis factors associated with outcome are

gathered into the International Prognosis Index (IPI) but in the last

years some other factors have been investigated in order to

discriminate high‐risk subgroups of patients. The red cell distribution

width (RDW) has been studied as a possible prognostic factor in

DLBCL and other malignancies. Our aim was to analyze if elevated

RDW (>15%) may help to differentiate patients affected by DLBCL

with a worse overall survival.

We studied 53 patients (male 51%, mean age 66 years, range 18‐91)

with a diagnosis of DLBCL treated in our institution between 2010 and

2015. At diagnosis, RDW was elevated in 21 patients (39,6%) and

lactate dehydrogenase was high in 32 cases (60,4%). Distributed ac-

cording to Ann Arbor, 3 patients (5,7%) were in stage 1, 8 patients in

stage 2 (15,1%), 14 patients in stage 3 (26,4%) and 28 patients in stage 4

(52,8%). Bone marrow inﬁltration was detected in 16 patients (30,2%)

and B symptoms were present in 52,8%. IPI score classiﬁed 13 cases

(24,5%) as low‐risk, 11 (20,8%) low‐intermediate, 17 (32,1%) high‐

intermediate, and 12 (22,6%) as high‐risk. After ﬁrst line treatment,

complete remission (CR) was achieved in 64,2% of patients. CR rate

was 71% in patients with normal RDW and 59% in those with elevated

RDW, but this difference was not signiﬁcant (p =0,53). Patients with a

high RDW value (>15%) at diagnosis showed a worse overall survival

(p =0,01; Figure 1a) and a worse progression‐free survival to ﬁrst line

treatment (p =0,039; Figure 1b), when compared with those with a

normal RDW. Multivariate analysis conﬁrmed the independent pre-

dictive value of RDW for overall survival.

This is one of several studies that shows that RDW value at

diagnosis constitute a simple, cheap and affordable prognostic factor

in DLBCL. What exactly RDW reﬂects in this disease (as well as in

other entities where its prognostic importance has been demon-

strated), from a pathophysiological point of view, requires a more in‐

depth investigation.

Keywords: Diagnostic and Prognostic Biomarkers

No conﬂicts of interests pertinent to the abstract.

339 |NEUTROPHIL/LYMPHOCYTE RATIO AND MONOCYTE/

LYMPHOCYTE RATIO – PROMISING PROGNOSTIC BIOMARKERS

IN DIFFUSE LARGE B‐CELL LYMPHOMA?

C. Afonso

, S. Duarte

, B. Marques

, C. Barros Lima

, D. Neves

, A.

Lai

, M. J. Julião

, L. Ruzickova

, A. Roque

, J. P. Carda

, M. Gomes

A. Cipriano

, A. I. Espadana

Centro Hospitalar e Universitário de Coimbra, Clinical Hematology

Department, Coimbra, Portugal,

Centro Hospitalar e Universitário de

Coimbra, Pathology Department, Coimbra, Portugal

Background: During the past few years, several studies have tried to

identify biomarkers that could improve the current prognostic

assessment in diffuse large B‐cell lymphoma (DLBCL). Neutrophil/

Lymphocyte Ratio (NLR) and Monocyte/ Lymphocyte Ratio (MLR) are

simple, cost‐effective and readily available tools that may have a

prognostic signiﬁcance in DLBCL treated with R‐CHOP, according to

some studies.

Aims: To determine the prognostic value of NLR and MLR in DLBCL,

Not Otherwise Speciﬁed (NOS).

Methods: Retrospective analysis of 335 patients (pts) with newly

diagnosed DLBCL, NOS, between 2010 and 2019 in a tertiary centre;

only those treated with R‐CHOP(‐like) regimens were included. The

receiver operating characteristic curve (ROC) method was applied in

resetting the optimal cut‐off point for predicting progression or

428

SUPPLEMENT ABSTRACTS

death from any cause. Variables with p <0.1 in univariate analysis

(UVA) were included in multivariate analysis (MVA).

Results: A total of 208 pts were identiﬁed. The median age at diag-

nosis was 63 years (24‐79), and 56% of pts were male. Stage III/IV

disease was present in 71% of pts, and B symptoms in 48%. Most pts

(78%) had extranodal involvement, and 51% presented lactate de-

hydrogenase (LDH) above the upper limit of normal (ULN). According

to the International Prognostic Index (IPI), 40% of pts were high‐

intermediate (HI) or high risk. The optimal cut‐off value for NLR

was 3.09 (sensitivity 70.2% and speciﬁcity 50.4%; AUC 0.597,

p=0.019) and for MLR was 0.45 (sensitivity 64.3% and speciﬁcity

60.0%; AUC 0.624, p =0.003).

Higher NLR was signiﬁcantly associated with Ann Arbor (AA) stage

III/IV (p =0.005), B symptoms (p =0.003), LDH>ULN (p <0.001),

and IPI HI/high risk (p <0.001). Similarly, a higher MLR was signiﬁ-

cantly associated with higher incidence rate of B symptoms

(p <0.001) and LDH>ULN (P <0.001).

Considering the whole cohort, the 5‐year progression‐free survival

(PFS) was 59.2%, and the 5‐year overall survival (OS) was 68.9%. In

the UVA, age >65 (PFS, p =0.005, HR 1.85; OS, p <0.001, HR 2.73)

AA stage III/IV (PFS, p =0.001, HR 2.67; OS, p =0.003, HR 2.77), IPI

HI/high risk (PFS, p =0.005, HR 1.88; OS, p =0.004, HR 2.11),

LDH>ULN (PFS, p =0.010, HR 1.77; OS, p =0.043, HR 1.66), NLR

≥3.09 (PFS, p =0.003, HR 2.02; OS, p =0.010, HR 2.01) and

MLR≥0.45 (PFS, p =0.001, HR 2.08; OS, p =0.018, HR 1.83) were

signiﬁcantly associated with shorter PFS and OS. However, in the

MVA, the NLR and MLR did not retain a statistically signiﬁcant as-

sociation with PFS and OS.

Conclusions: In our cohort, elevated NLR and MLR were signiﬁcantly

associated with several high risk clinical features, as well as poorer

outcome. However, their association with PFS and OS lost its sig-

niﬁcance in MVA, suggesting that both ratios are not independent

prognostic factors in DLBCL pts undergoing treatment with R‐CHOP.

Therefore, the role of NLR and MLR in DLBCL may require further

validation before widespread clinical use.

Keywords: Diagnostic and Prognostic Biomarkers

No conﬂicts of interests pertinent to the abstract.

340 |RETROSPECTIVE APPLICATION OF THE CNS‐IPI TO A

POPULATION OF DLBCL PATIENTS TREATED WITH IT MTX

PROPHYLAXIS: A SINGLE CENTRE ANALYSIS.

C. Flynn

, M. McCarthy

, T. Harvey

, R. Flavin

, B. Dunne

, S. Sukor

C. Grant

St James's Hospital, Medical Oncology, Dublin, Ireland,

St James's

Hospital, Pathology, Dublin, Ireland

Introduction: Central nervous system (CNS) relapse in diffuse large

B‐cell lymphoma (DLBCL) is associated with a poor prognosis, with

a median survival of four months. There was no reproducible model

to estimate CNS relapse risk until the “CNS‐IPI” was reported

(Schmitz et al. JCO 2016). This uses six criteria to categorise pa-

tients with DLBCL into low, intermediate, and high risk for CNS

relapse with rates of 0.6%, 3.4% and 10.2% respectively at two

years.

CNS prophylaxis is given to those felt to be at high risk of CNS

relapse, however evidence for its use in the rituximab era is limited.

Traditionally, intrathecal methotrexate (IT MTX) has been used as

CNS prophylaxis, with rationale extrapolated from acute lympho-

blastic leukaemia studies. High‐dose intravenous methotrexate (HD

MTX) has emerged as an alternative prophylactic strategy, with some

data to suggest improved outcomes. However, some recent studies

have failed to show beneﬁt of either intervention.

Methods: A retrospective analysis was carried out on oncology pa-

tients treated with IT MTX as CNS prophylaxis for DLBCL in St

James's Hospital over a 7‐year period (2013‐2019). Double and

triple‐hit lymphomas were excluded. Baseline clinical and histo-

pathological data was recorded, and CNS‐IPI scores were retro-

spectively applied. Survival data was analysed by the Kaplan‐Meier

method.

Results: 37 patients were included with a median follow‐up of 48

months (r =3‐87). Median age was 67 (r =34‐86) and 62% were

male. 4 patients (11%) were HIV‐positive. 16 patients (43%) had GC

subtype and 19 (51%) had non‐GC subtype disease. Most patients

were treated with R‐CHOP (54%) or R‐EPOCH (43%) chemo-

therapy. 6 patients were CD5+. 84% had stage III‐IV disease, and

59.5% had an elevated LDH at diagnosis. When CNS‐IPI was

applied, 18 (48.6%), 14 (37.8%) and 5 patients (13.5%) were clas-

siﬁed as high, intermediate, and low risk for CNS relapse

respectively.

94.6% (n =35) achieved complete response with an overall survival

(OS) of 86.4% and progression‐free survival of 81%. OS was 77.8%,

92.8% and 100% in the high, intermediate, and low‐risk groups

respectively. 1 of 37 patients (2.7%) developed CNS relapse.

When divided by CNS‐IPI, CNS relapse rate was 0% for low

and intermediate‐risk groups, and 5.6% for high‐risk patients

(p =0.149). Five patients (13.5%) developed signiﬁcant toxicity with

IT MTX. Two patients developed subdural haematomas, two

developed severe back pain, and one developed a refractory

headache.

Conclusion: CNS relapse rate in this cohort is low, particularly for

low and intermediate‐risk CNS‐IPI groups, which both had rates of

0%. This is in keeping with other studies which have demonstrated

lower CNS relapse rates in the rituximab era. More studies are

required to further elucidate risk factors for CNS relapse and to

establish the role of CNS prophylaxis in the rituximab era. This is

particularly important now there has been a shift towards HD MTX,

which has much higher rates of toxicity.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Prevention and

Cancer Interception

No conﬂicts of interests pertinent to the abstract.

SUPPLEMENT ABSTRACTS

429

341 |CLINICAL RISK SCORES IN DIFFUSE LARGE B CELL

LYMPHOMA ‐IS THERE STILL ROOM FOR IMPROVEMENT?

D. S. Neves

, S. Duarte

, C. Afonso

, B. Marques

, C. B. Lima

, A. Lai

A. Roque

, L. Ruzickova

, J. Carda

, M. Gomes

, A. Cipriano

, A.

Espadana

Centro Hospitalar e Universitário de Coimbra, Clinical Hematology

Department, Coimbra, Portugal,

Centro Hospitalar e Universitário de

Coimbra, Pathology Department, Coimbra, Portugal

Introduction: Diffuse Large B Cell Lymphoma (DLBCL), the most

common non‐Hodgkin lymphoma, despite having a standard of care

treatment (R‐CHOP), has heterogeneous clinical and molecular fea-

tures. The widely used risk scores are based on clinical and

biochemical parameters, which can underestimate the impact of

genomic signatures. Recognizing the score with better discriminative

power might be important to better identify the high‐risk group, for

whom R‐CHOP may be insufﬁcient. We aim to compare the prog-

nostic efﬁcacy of scores used in DLBCL: IPI (International Prognostic

Index), NCCN‐IPI and GELTAMO‐IPI.

Methods: Retrospective analysis of 335 HIV negative pts with newly

DLBCL, uniformly treated with R‐CHOP (like) between 2010‐2019.

Pts treated with palliative intent were excluded. The discrimination

value for progression‐free survival (PFS) and overall survival (OS)

was performed using log‐rank statistics. Multivariate analysis

included variables statistically signiﬁcant (p‐value <0.05) in univar-

iate analysis, not included in each score calculation. Stratiﬁed models

for each risk scores were compared using Akaike's information cri-

terion (AIC) and concordance index (C‐index).

Results: A total of 228 pts was eligible, mainly male (54%, n =123),

median age 64 years (22‐87). At diagnosis 69.3% (n =158) were

stage III/IV, 40.4% had ≥2 extranodal sites and 28.6% had bulky

disease (n =62/217).

PFS at 5 years for low (L), low‐intermediate (LI), high‐

intermediate (HI), and high‐risk (H) was 79.9/58.7/51.6/39.5% for

IPI (HR =1.5; p <0.001),74.7/70/55.3/28.6% for NCCN‐IPI

(HR =1.7; p <0.001) and 93.3/67.3/35/25.7% for GELTAMO‐IPI

(HR =1.9; p <0.001). After multivariate analysis, the scores

remained discriminant for PFS (HR =1.6; p =0.002; HR =1.7;

p=0.007; HR =1.7; p =0.001, respectively).

OS at 5 years for L, LI, HI, and H was 86/67.9/63.7/41.6% for IPI

(HR =1.6; p <0.001), 85.7/80/62/30.8% for NCCN‐IPI (HR =2.2;

p<0.001) and 93.3/76/44/37.5% for GELTAMO‐IPI (HR =2;

p<0.001).

For PFS, GELTAMO‐IPI has a substantial improvement in the

model ﬁtness (AIC 633/950/951 for GELTAMO‐IPI, IPI and NCCN‐

IPI, respectively) and a slightly better discrimination between short

and long survival times (C‐index 0.6426/0.6378/0.6095) compared to

the other scores, that also have an acceptable discrimination capac-

ity. OS showed concordant results (AIC 480/733/725 respectively;

C‐index 0.6916/0.6543/0.6573).

Thus, for both PFS and OS, GELTAMO‐IPI cannot accurately

identify very high‐risk pts.

Conclusions: Our results showed that GELTAMO‐IPI has only a

marginal performance advantage in discriminating risk groups, when

compared with IPI and NCCN‐IPI. However, none of them precisely

identify the high‐risk DLBCL pts, which highlight the need to develop

new prognostic scoring systems with improved capacity of risk

stratiﬁcation, providing the selection of those in need of novel

therapies.

Keywords: Diagnostic and Prognostic Biomarkers

No conﬂicts of interests pertinent to the abstract.

342 |DIFFUSE LARGE B CELL AND HIGH‐GRADE LYMPHOMAS:

THERAPEUTIC RESPONSE ‐DESCRIPTIVE RETROSPECTIVE

UNICENTRIC ANALYSIS

C. L. Pedrosa

, B. Castro

, R. P. Peixeiro

, L. Regadas

, R. Coutinho

R. Cabral

, J. Coutinho

Centro Hospitalar Universitário do Porto, Clinical Hematology, Porto,

Portugal

Introduction: Relapse/refractoriness (R/R) in diffuse large B cell

(DLBCL) and high grade (HGL) lymphomas affects more than 1/3 of

the patients, constituting an unfavorable prognosis and therapeutic

challenge.

Methods: This single‐center retrospective analysis aims to describe

the response to 1st line and salvage treatment of patients with

DLBCL and HGL diagnosed between 2015‐2019.

Results: 125 patients were diagnosed. Of these, 58 did not undergo

curative chemoimmunotherapy (R‐ChT) (median age 76 years;

ECOG ≥2 in 40%). We analyzed 67 patients; 55% men; median age

61 years (52% >60 years); ECOG ≥2 in 10%. 10% were HIV+. 81%

were DLBCL (NOS n =47), 13% HGL and 6% PTLD. 9% had pri-

mary extranodal involvement. The IPI score was calculated: low:

26%; low intermediate: 30%; high intermediate: 33%; high: 7%. The

R‐ChT schemes were R‐CHOP (n =44), R‐CNOP (n =10), R‐mini‐

CHOP (n =8), others (n =5). Dose adjustment /replacement of

anthracycline was performed in patients who were more than 60

years‐old, had high ECOG and/or comorbidities. 25% underwent

radiotherapy in the ﬁrst line treatment. 67% achieved complete

response (CR). The rate of R/R was 44.8% (relapse n =10;

refractoriness n =20). The median time to relapse was 10.5

months. At diagnosis, these patients had an IPI ≥2 in 87%, bulky

disease in 50% and 73% were CD5+. The activated B cell subtype

was identiﬁed in 33%. Five were double/triple‐expressers. Two

double‐hit lymphomas were identiﬁed. Twenty‐four patients un-

derwent a second line treatment. Only 29% reached CR (R‐ICE

n=3; R‐DHAP n =2; R‐Bendamustine n =1; CVP n =1). Four

patients underwent autotransplantation. Eight received ≥3 thera-

peutic lines. The median follow‐up was 19.3 months. The mortality

rate was 36% (52% of the total diagnosed). 71% died from disease

progression. The median overall survival of patients undergoing 1st

R‐ChT was 39 months.

430

SUPPLEMENT ABSTRACTS

Conclusion: Due to advanced age, comorbidities and/or high ECOG,

only 54% of patients with DLBCL and HGL were chosen for curative

R‐ChT. Only 9 are alive with a follow‐up time ≥18 months. The

studied patients presented characteristics, risk factors and unsatis-

factory responses to R‐ChT in R/R, as described. Thus, we emphasize

the adjusted selection of treatment strategies as a way to improve

the prognosis.

EA – previously submitted to regional or national meetings (up to

1000 attendees).

Keywords: Diagnostic and Prognostic Biomarkers, Aggressive B‐cell

non‐Hodgkin lymphoma, Chemotherapy

No conﬂicts of interests pertinent to the abstract.

343 |THE PRIMARY RESULTS OF THE EFFICACY OF UPFRONT

HIGH‐DOSE CHEMOTHERAPY WITH AUTOLOGOUS STEM CELL

TRANSPLANTATION FOR IV STAGE AND HIGH‐INTERMEDIATE/

HIGH IPI RISK GROUPS DLBCL

A. Koviazin

, L. Filatova

, I. Zyuzgin

, A. Artemyeva

, M. Motalkina

Y. Chudinovskikh

, S. Volchenkov

, I. Poliatskin

, E. Cherkasova

S. Shalaev

, I. Ishmatova

, A. Zverkova

, M. Ghanzin

, Y. Oleinik

D. Burda

, T. Semiglazova

NMRC of Oncology named after N.N.Petrov of MoH of Russia,

Department of Hematology and Chemotherapy with Intensive Care unit,

Saint‐Petersburg, Russian Federation,

NMRC of Oncology named after N.

N.Petrov of MoH of Russia, Department of Pathology, Saint‐Petersburg,

Russian Federation,

NMRC of Oncology named after N.N.Petrov of MoH

of Russia, Department of Innovative Methods in Therapeutic Oncology

and Rehabilitation, Saint‐Petersburg, Russian Federation

Introduction: Approximately 30% of patients(pts) with IPI ≥2 diffuse

large B‐cell lymphoma (DLBCL) have relapsed after the frontline reg-

imens. High‐dose chemotherapy (HDCT) with autologous hematopoi-

etic stem cell transplantation (auto‐HSCT) in the ﬁrst remission can be

an effective option to decrease the relapse rate in these patients.

Methods: 108 pts, which ﬁt the following criteria: DLBCL NOS, age

18‐65, stage IV, IPI ≥2, treated by x6 CHOP/EPOCH/H‐CVAD +R in

frontline from 01.01.2010 to 31.12.2019 at NMRC of Oncology

named after N.N.Petrov of MoH of Russia were retrospectively

analyzed.

If patients achieved a complete response (CR) or partial response

(PR) after frontline, they enrolled into upfront HDCT with auto‐HSCT

group (Group 1, n =38) or follow‐up group (Group 2, n =70) based

on multidisciplinary team (MDT) decision. More than one extranodal

involvement (extra) had 78.9% (30/38) pts in Group 1 and 75.7% (53/

70) pts in Group 2. MYC and BCL‐2 coexpression were analyzed in

84% (32/38) pts samples in Group 1 and 65.7% (46/70) in Group 2

and were present (DEL) in 46.8% (15/32) pts samples in Group 1 and

28.2% (13/46) in Group 2.

A Fisher's exact test (Fisher's) with a two‐tailed p‐value was used to

determine the difference in relapse rate between Group 1 and Group

2. A McNemar test (McNemar) with Yates correction (Yates) was

used to determine the difference in responses in Group 1 after

upfront HDCT with auto‐HSCT.

Results: CR after induction was achieved in 61% (23/38) pts in Group

1. After HDCT with auto‐HSCT, the CR rate increased to 97% (37/

38) and continues for a median of 30 (7‐56) months in 87% (33/38)

pts. It demonstrates the statistically signiﬁcant difference – McNe-

mar =11.27 (p ≤0.001) and with Yates =10.42 (p =0.002). Early

relapses (ER) were diagnosed in 6% (2/38) pts, late relapses (LR) in

6% (2/38) pts. HDCT associated mortality (infection) has occurred in

3% (1/38) pts.

Group 2 pts achieved CR in 81% (57/70) cases. CR continues for

median 38 (8‐73) months in 67% (47/70) pts. Group 2 pts have 14%

(14/70) ER and 6% (4/70) LR. No treatment‐related deaths occurred.

ER rate was higher in Group 2 (p =0.048). In Group 2 pts with ˃1

extra experienced more ER (Group 1 – 3.3% (1/30), Group 2 – 20.8%

(11/53), p =0.048). DEL was signiﬁcant in terms of both ER (Group 1

– 6.6% (1/15), Group 2 – 38.4% (6/13), p =0.029) and all relapses

(AR) (Group 1 – 6.6% (1/15), Group 2 – 53.8% (7/13), p =0.011). The

ER rate in Group 2 was signiﬁcantly higher if the combination of ˃1

extra and DEL were present (Group 1 – 9.1% (1/11), Group 2 – 60%

(6/10), p =0.0237).

Conclusions:

1. CR rate signiﬁcantly increased in DLBCL NOS pts with stage

IV, IPI ≥2 treated by upfront HDCT with auto‐HSCT.

2. ER rate signiﬁcantly decreases in DLBCL NOS with stage IV,

IPI ≥2 after upfront HDCT with auto‐HSCT pts if ˃1 extranodal site,

DEL, and combination of these factors are present; AR rate decreases

if pts only have DEL.

EA – previously submitted to EHA 2021.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Stem Cell

Transplant

No conﬂicts of interests pertinent to the abstract.

344 |REAL WORLD APPLICATION OF R‐CHOP FOR DIFFUSE

LARGE B‐CELL LYMPHOMA (DLBCL) IN BELGIUM AND ITS

IMPACT ON SURVIVAL.

W. Daneels

, M. Rosskamp

, G. Macq

, E. Saadoon

, A. Degeyndt

, F.

Offner

, H. Antoine‐Poirel

Ghent University Hospital, Department of Hematology, Ghent, Belgium,

Belgian Cancer Registry, Research Department, Brussels, Belgium

Introduction: There is a growing interest for real‐world population

studies to compare the results of randomized clinical trials in an

unselected population, including the elderly and patients with

comorbidities or other malignancies. We evaluated the patterns of

SUPPLEMENT ABSTRACTS

431

care for DLBCL in Belgium including the use of R‐CHOP and radio-

therapy (RT).

Methods: Coded data of all adult (≥20 year) DLBCL diagnosed 2013‐

2015 (n =1,888) were obtained via the Belgian Cancer Registry. Vital

status was available until July 1

2019. Data was extracted from

pathology reports (IHC & FISH biomarkers) and oncological care

programs (performance status & staging). Chemotherapy, stem cell

transplantation and RT were inferred from health insurance data of

reimbursed treatments. An in‐house algorithm identiﬁed chemo-

therapy regimens (e.g. R‐CHOP/R‐DHAP), the number of cycles and

intervals, switches between regimens and refractoriness. Inherent

limitations apply due to treatment inference and some missing

prognostic markers.

Results: Systemic ﬁrst‐line treatment was started in 85%, decreasing

with age to only 46%. Treatments varied signiﬁcantly by age.

Anthracycline based regimens were most frequently used (76%),

even in the elderly (45%). R‐CHOP was used in 1163/1596(73%) of

treated patients. We could not discriminate R‐CHOP from R‐

miniCHOP, the preferred regimen in the elderly.

Survival 2‐year OS of all patients was 63% with a clear inﬂuence of

age (31‐84%) and linked to ﬁrst‐line treatments: standard R‐CHOP

(85%), other anthracycline(66%), incomplete R‐CHOP (55%), non‐

anthracycline (44%), only RT (43%) and no chemo/RT (15%).

Standard R‐CHOP The median [IQR] cycle interval was 21 [21‐22]

days, consistent across age groups. The median [IQR] number of

cycles was 6 [4‐7] up to the age of 89, except for the age group 20‐39

with 4 [2‐6] cycles. In 14% of patients aged 85‐89, and 5 patients

aged >90, ≥6 cycles were given. In multivariable analyses for OS,

standard R‐CHOP was associated with a HR of 0,38 (0,30‐0,48).

Incomplete R‐CHOP ±RT In 337/1163 cases (29%) <6 cycles (or <4 if

Ann Arbor =I) were given (<4 cycles in 195/337 and 4‐5 in 142/337).

Reasons could not be determined but include early death, resistance,

local disease and part of extended and/or non‐reimbursed regimens. In

40/337 a second line regimen was started. RT was applied after <4 and

4‐5 cycles in 39 and 30 patients. 5‐year OS ranged between 20‐78%

(Fig. 1) with RT groups and primary refractory cases associated with

the highest and lowest OS respectively.

Conclusions: For DLBCL in the Belgian real‐world setting, R‐CHOP‐21

is the primary ﬁrst‐line treatment and completing 6 cycles is achieved

in the majority of patients, including the elderly. In 29% of cases <6

cycles were given, but those receiving consolidative radiotherapy

maintained a favorable OS, conﬁrming the curative potential of less

immunochemotherapy followed by radiotherapy in selected patients.

Keywords: Diagnostic and Prognostic Biomarkers, Aggressive B‐cell

non‐Hodgkin lymphoma, Combination Therapies

No conﬂicts of interests pertinent to the abstract.

345 |OUTCOMES OF DIFFUSE LARGE B CELL LYMPHOMA

PATIENTS TREATED IN A BRAZILIAN PUBLIC CANCER CENTER –

A REAL WORLD EXPERIENCE OF 809 PATIENTS IN A 11 YEAR

PERIOD.

M. Bellesso

, J. Pereira

, L. Bassolli

, A. N. R. Abdo

, W. F. S. Junior

R. D. Velasques

, A. A. G. S. Brandão

, R. B. Melo

, L. A. P. C. Lage

F. V. R. Maciel

, H. Visnadi

, R. Shcolnik

, G. Dufﬂes

, J. I. Beraldo

A. Duran

, I. C. Soares

, S. A. C. Siqueira

, V. Rocha

Instituto do Câncer do Estado de São Paulo, Hematology, São Paulo,

Brazil,

Instituto do Câncer do Estado de São Paulo, Pathology, São Paulo,

Brazil,

Hospital das Clínicas da Faculdade de Medicina da Universidade

de São Paulo, Pathology, São Paulo, Brazil

432

SUPPLEMENT ABSTRACTS

Introduction: Diffuse Large B cell Lymphoma (DLBCL) is a heterog-

enous disease. Factors based on clinical and simple laboratory pa-

rameters are known and widely used such as the Revised

International Prognostic Index (R‐IPI) and National Comprehensive

Cancer Network ‐IPI (NCCN‐IPI). Thus, we aimed to report real

world experience of outcomes of DLBCL lymphoma patients treated

in a Brazilian public cancer institution and evaluate the prognostic

role of R‐IPI and NCCN‐IPI.

Methods: This is an unicenter registry study including DLBCL pa-

tients in a public tertiary cancer center in São Paulo city, Brazil.

Consecutive patients diagnosed from 01/2009 to 12/2020, aged

≥18 years, treated with R‐CHOP or R‐CHOP like regimens were

included. Patients with AIDS related lymphoma, or primary medi-

astinal lymphoma were excluded. Patients were stratiﬁed by R‐IPI

and NCCN‐IPI. Progression free survival (PFS) and overall survival

(OS) were estimated by Kaplan‐Meier curve. All variables studied

showed less than 10% of missing data and valid percent were

reported.

Results: We analyzed 809 patients who met the eligibility criteria.

416 (51.5%) were males. Median age was 59.9 (IQR 44.9 – 68.2)

years, 271 (31.3%) presented ECOG ≥2, 561 (74.9%) advanced

disease, 695 (87.4%) elevated LDH and 638 (82.2%) extra‐nodal

involvement. Patients were classiﬁed according to prognostic

scores, 411 (50.8%) were considered poor prognosis by R‐IPI (≥3 pts)

and 465 (57.5%) were classiﬁed as high risk or high‐intermediate

NCCN‐IPI (≥4 pts).

Response status after ﬁrst line treatment was available in 678 pa-

tients, 456 (67.2%) achieved CR. Median follow‐up 4.6 ‐95CI (4.2‐

5.0) years.

Estimate 4 year OS and PFS for the entire patient cohort were 65.3%

and 60.4%, respectively.

Patients identiﬁed as very good (0), good (1‐2) and poor (≥3)prog-

nosis by R‐IPI presented 4‐year OS 94%, 75.2% and 52.8%, p <

0.0001, and 4‐years PFS was: 94%, 68.8%, 48.2%, respectively, p <

0.001. (Figure)

Estimate 4 year OS and PFS according to NCCN‐IPI: low (0‐1), low‐

intermediate (2‐3), high‐intermediate (4‐5) and high risk were

respectively 95.5%, 75.4%, 60.3%, 35.9%, p <0.0001) and 93.7%,

69.7%, 54.7% and 33.3%, p <0.001, respectively. (Figure).

Conclusion: In this large cohort of DLBCL patients diagnosed and

treated in a public University center in Brazil, we have observed a

higher frequency of patients with unfavorable prognosis 57.5%

SUPPLEMENT ABSTRACTS

433

NCCN‐IPI (≥4 pts) and 50.8% R‐IPI (≥3 pts) when compared to data

published in the literature 39% and 45%, respectively. OS and PFS

are comparable to data of literature according to NCCN‐IPI and R‐IPI

scores. Both scores discriminate groups of patients with DLBCL in a

middle‐income country and may be used to study further strategies

to improve outcomes of patients with poor prognosis.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

346 |TREATMENT AND OUTCOMES OF PATIENTS WITH

POST‐TRANSPLANT LYMPHOPROLIFERATIVE DISORDER: A

SINGLE CENTRE EXPERIENCE

C. de Ramon Ortiz

, K. Samii

, K. Hadaya

, A. Avram

, Y. Beauverd

Geneva University Hospitals, Haematology Division, Department of

Oncology, Geneva, Switzerland,

Geneva University Hospitals, Nephrology

Division, Department of Medicine, Geneva, Switzerland,

Geneva

University Hospitals, Pathology Division, Department of Genetics and

Laboratory Medicine, Geneva, Switzerland

Background: Post‐transplant lymphoproliferative disorder (PTLD) is

a potentially severe complication that takes place after solid (SOT)

or allogeneic hematopoietic stem cell transplant (HSCT). To

date, its treatment is not standardised, and choice of therapy de-

pends on histology and clinical presentation. We report the results

of a 10 years (yrs) single‐centre retrospective study assessing

presentation, treatment patterns and outcomes of patients with

PTLD.

Methods: Patients ≥18 yrs diagnosed with a PTLD between 2011

and 2020 were identiﬁed in our medical records database. Primary

endpoints were 2 yrs progression‐free survival (PFS), overall sur-

vival (OS), and relapse incidence (RI) estimated with Kaplan‐Meier

and comparison between HSCT and SOT group. Secondary end-

points were the overall response rate (ORR) and complete

response rate (CRR) at the end of ﬁrst line treatment. Exploratory

univariate analysis for OS, PFS were investigated using Cox

regression model.

Results: A total of 38 patients were included. Patient's characteristics

and treatment strategies are presented in Table 1. HCST patients

present a signiﬁcantly higher prevalence of early onset PTLD (90% in

HCST group vs 0% in SOT group), positive EBER immunostaining

(100% vs 22%) and detectable Epstein Bar Virus (EBV) viremia (100%

vs 28%), higher LDH (95% vs 66%), IPI score 3‐5 (50% vs 33%), with a

remarkable higher use of single agent Rituximab as ﬁrst line therapy

in HSCT patients (65% vs 11%). 2 yrs PFS was 54% (95%CI 36‐73%)

for the entire cohort, 42% (95%CI 16‐67%) for HSCT and 67% (95%

CI 41‐92) for SOT; p‐value 0.23. 2 yrs OS was 53% (95%CI 41‐66%)

for the entire cohort, 42% (95% CI 16‐67%) for HSCT and 65% (95%

CI 39‐91) for SOT; p‐value 0.22. 2 yrs RI was 9% (95% CI: 0‐21%) for

the entire cohort, 0% for HSCT and 14% (95% CI 0‐33%) for SOT; p‐

value 0.26. ORR and CRR were similar with 82% (75% for HSCT, 89%

for SOT, p‐value 0.41) at the end of ﬁrst line treatment. High IPI

score (3‐5) demonstrated worse outcome for PFS (HR: 2.5, 95%CI

0.95‐6.65, p‐value 0.063) and OS (HR: 2.4, 95%CI 0.92‐6.48, p‐value

0.07). A total of 17 patients died during follow‐up: 7 during ﬁrst line

treatment of PTLD complication and multiorgan failure (5 in the

HSCT group, 2 in the SOT group; p‐value 0.41); 2 following PTLD

relapse after ﬁrst CR (0 in the HSCT group, 2 in the SOT group;

p‐value 0.22).

TABLE 1Patient's baseline characteristics and ﬁrst line

therapy

434

SUPPLEMENT ABSTRACTS

Conclusion: This study highlights the clinical and therapeutic het-

erogeneity of PTLD. We observe a worse 2 yrs PFS and OS in patients

with high IPI score. CRR after ﬁrst line treatment is almost 90% in the

SOT group, with a 2 yrs OS of 53%, consistent with previous studies

(Trappe et al, JCO 2017). The higher mortality observed in the HSCT

group suggests that serious treatment‐related toxicities remain main

challenges in this fragile patient population. New tools to guide

therapeutic approaches in PTLD patients are warranted.

Keywords: Lymphoid Cancers ‐Other

No conﬂicts of interests pertinent to the abstract.

347 |DOUBLE‐HIT (DHL) AND TRIPLE‐HIT LYMPHOMAS (THL):

REAL LIFE EXPERIENCE OF 46 CONSECUTIVE PATIENTS FROM A

SINGLE INSTITUTION IN SPAIN

E. M. Gonzalez Barca

, J. M. Sancho

, J. Rovira

, N. Kelleher

M. Kara

, M. Encuentra

, E. Domingo Domenech

, A. C. Oliveira

J. M. Ribera

, L. Escoda

, A. Sureda

Institut Català d'Oncologia, Hospital Duran i Reynals, Institut

d'Investigació Biomèdica de Bellvitge (IDIBELL), Universitat de Barcelona,

Hematology, L'Hospitalet de Llobregat, Barcelona., Spain,

Institut Català

d'Oncologia, Hospital Germans Trias i Pujol, Institut de Recerca contra la

Leucèmia Josep Carreras, Universitat Autònoma de Barcelona, Badalona,

Hematology, Barcelona, Spain,

Institut Català d'Oncologia, Hospital

Universitari Joan XXIII, Hematology, Tarragona, Spain,

Institut Català

d'Oncologia, Institut d'Investigació Biomèdica de Girona (IDIBGI),

Universitat de Girona, Hematology, Girona, Spain

Introduction: Double‐hit (DHL) and triple‐hit lymphomas (THL) have

been described in the 2016 revised WHO classiﬁcation, they repre-

sent a new entity called high‐grade B‐cell lymphoma (HGBCL) with

rearrangements of MYC and BCL2 and/or BCL6. DHL/THL have been

described as clinically aggressive with a high risk of central nervous

system (CNS) and extranodal organ involvement and with a worse

prognosis when treated with standard R‐CHOP.

Patients and Methods: Retrospective study of 46 consecutive pa-

tients with HGBCL with DHL/THL, diagnosed from March 2015 to

November 2020, in centers of the Catalan Institute of Oncology in

Spain. Survival curves were plotted by the Kaplan‐Meier method.

Comparison analysis for survival was performed with the log‐rank

test.

Results: Clinical characteristics at diagnosis are shown in the table.

Of note, 98% presented extranodal disease: bone marrow (n =11),

bone (n =11), kidney/adrenal (n =10), lung (n =9), gastrointestinal

(n =8), soft tissue (n =8), and CNS involvement (n =2). Three pa-

tients did not receive ﬁrst line treatment (2 due to old age and 1

patient died few days after diagnosis). Among the 43 patients

treated, regimens used were: R‐CHOP: 22 (51%) patients (2 of them

R‐miniCHOP), DA‐EPOCH‐R: 14 (33%), intensive regimen used in

Burkitt lymphoma: 4 (9%), R‐GEMOX: 1 patient, radiotherapy:1 pa-

tient, unknown:1. CNS treatment or prophylaxis consisted in high‐

dose intravenous methotrexate in 9 patients, and intrathecal meth-

otrexate in 5. No patient received stem‐cell transplant in ﬁrst line.

Overall response was 67%; complete remission (CR): 23 (53%), par-

tial remission (PR): 6 (14%), progression: 12 (28%) and 2 patients

were not evaluable. CR was 45% after R‐CHOP vs 67% with inten-

sive regimens. With a median follow‐up for the alive patients of 41

months, 23 (50%) patients died, 18 (39%) due to progressive lym-

phoma, 3 due to septic shock, 1 gastric carcinoma and 1 severe

bronchiolitis. There were not CNS relapses. Progression‐free survival

(PFS) and overall survival (OS) at 3 years of the 46 patients were 42%

(95% CI 26‐57) and 47% (95% CI 38‐62), respectively. PFS at 3 years

FIGURE 1 Survival curves for PFS and OS

SUPPLEMENT ABSTRACTS

435

of patients treated with R‐CHOP versus patients treated with

intensive regimens was 40% (95% CI 17‐63) versus 51% (95%CI 25‐

77), and OS at 3 years was 47% (95%CI 24‐69) versus 56% (95%CI

29‐83), without statistical signiﬁcant differences.

Conclusions: Almost all patients with HGBCL DHL/THL have extra-

nodal involvement at diagnosis and have other factors of poor

prognosis as previously described. Survival of the whole series is

poor. Nevertheless, although this is a small series, our data suggest

that higher therapeutic intensity could improve outcomes, but PFS

and OS of patients treated with R‐CHOP are better than in other

older retrospective studies. Therefore, a prospective randomized trial

is mandatory to clarify the role of intensive treatments in these pa-

tients.

FIGURE 1: Progression free survival of patients with high grade

lymphoma double.triple hit according to the ﬁrst line of treatment

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

348 |PRIMARY ORAL CAVITY LYMPHOMAS (POCL).

FEATURES OF 67 PATIENTS TREATED AT A SINGLE INSTITUTION

IN ARGENTINA

A. C. Jozami

, J. Arriola

, M. Marquez

, P. Luchetta

, I. Annetta

, N.

Diaz Velez

, V. Courreges

Oncology Hospital Maria Curie, Hematology, Buenos Aires, Argentina

Introduction: POCL represents 2‐8% of head and neck cancers. They

are characterized by their locoregional growth, low speciﬁcity of the

initial symptoms, and a predominance of aggressive histology.

Objective: To describe the characteristics of 67 patients with POCL.

Methods: the medical records and pathological anatomy studies of

patients admitted to the Hematology and Stomatology Services be-

tween the years 1996 and 2020 were evaluated.

Results: Ages: 19 to 88 years old, mean: 52; Masc. 44 and, Fem. 23.

436

SUPPLEMENT ABSTRACTS

Locations: lower jaw: 45% upper jaw 27%, base of tongue 13%, hard

palate 11%, ﬂoor of mouth 4%.

Histology: diffuse large B‐cell lymphoma (DLBCL): 52%, plasma-

blastic lymhoma (PBL): 24%, Follicular L 10%; PTCL NOS 7%; Burkitt

2%, NK/T‐cell L: 1,5%; MALT: 1,5%.

Stages: I: 66%, II: 25% IV: 9%, due to bone marrow inﬁltration

Increased LDH was observed in 33% of patients, β‐2microglobulin in

23%, and erythrocyte sedimentation in 23%.

Toxic antecedents: tobacco 30% and dental prostheses in 15%.

The initial symptoms were mobility and pain of the teeth, a rapidly

growing painful tumor in the jugal mucosa, and torpid oral ulcers.

HIV positive in 16 patients, 13 with PBL and 7 with DLBCL.

Treatment: it was carried out according to the histology, the patient's

PS, viral status and date of diagnosis.

The following regimes were used: CHOP, R‐CHOP, CVP; R‐CVP,

CAVPE, CNOP; DA‐EPOCH. In 15 patients there are no data on the

treatment carried out. 6 cases received local radiotherapy. CNS

prophylaxis with systemic methotrexate or intrathecal drugs was

performed according to the schedule chosen and the patient's risk

factors.

Responses: complete remission (CR) in 37%, partial remission (PR) in

16% of those who underwent treatment. There's no data in 22 pa-

tients due to loss of follow‐up.

Follow‐up time ranged from 1 year to 21 years.

Conclusions: In this series of patients, we observed most aggressive

histology, with fast growing locoregional tumors. Dental alterations

led to consultations with stomatologists, who took diagnostic bi-

opsies. There was a predominance of males, of B lineage, with middle

ages predominate. The IPI and the stage were not useful tools for the

evaluation of this pathology.

The association with HIV was observed in 28% of the cases, the

majority were PBL (68%), with increased DLBCL histology.

Favorable responses were observed, with prolonged survival in cases

with less aggressive histology.

Keywords: Extranodal non‐Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

349 |LONG TERM OUTCOMES OF EARLY STAGE PRIMARY

GASTRIC DIFFUSE LARGE B‐CELL LYMPHOMA: A MULTI‐CENTRE

RETROSPECTIVE STUDY

C. D. Smith

, Y. S. Chin

, S. A. Gupta

, S. R. Thompson

Prince of Wales Hospital, Department of Radiation Oncology, Randwick,

Australia,

St George Hospital, Department of Radiation Oncology,

Kogarah, Australia,

Calvary Mater Hospital, Department of Radiation

Oncology, Newcastle, Australia

Introduction: Early stage gastric disease is an uncommon sole

manifestation of diffuse large B‐cell lymphoma (DLBCL). Radio-

therapy (RT) can be used as part of a combined modality approach

in the treatment of early stage nodal DLBCL with excellent results.

In primary gastric DLBCL (PG‐DLBCL) the role of RT is not well

established and outcome data are lacking. We assessed the long

term outcomes of patients who received RT as part of their

curative‐intent treatment for early stage PG‐DLBCL at three NSW

institutions.

Methods: We carried out a retrospective, multi‐centre study of pa-

tients with PG‐DLBCL who had had treatment with radiotherapy

between 01/03/1999 and 29/05/2020. Eligible patients were age

≥18 years, received radiotherapy as part of combined modality

therapy, pathological diagnosis of DLBCL, Lugano stage I‐II. Time‐to‐

event outcomes were calculated from day 1 of treatment. Toxicities

were extracted from the medical chart per CTCAE v4.0

Results: There were 14 eligible patients. Median age was 75 years

(range 48 – 84). 10 (71%) patients had Lugano stage I disease, 4

(29%) had stage II. IPI at diagnosis was 0 in 14%, 1 in 57% and 2 in

29%. All patients previously received systemic therapy (11 R‐CHOP,

1 CHOP, 2 other). 5 (33.3%) patients were H pylori positive and all 5

received H Pylori eradication prior to treatment. Median prescribed

RT dose was 30.6Gy (range 30 – 45) in 17.5 fractions (range 15 ‐25).

10 (71%) treatments were delivered with 3D conformal technique, 4

(29%) with VMAT.

Thirteen patients had disease assessed prior to consolidation

RT: 10 (77%) achieved complete response, and 3 (23%) partial

response. During median follow up of 56 months (IQR 29‐70),

estimated 5 year local relapse free survival (LRFS), distant relapse

free survival (DRFS), and progression free survival (PFS) were 100%,

93% (95%CI; 80.3‐100), and 93% (95%CI; 80.3‐100), respectively.

There was 1 documented recurrence which was out of RT ﬁeld. 5

year OS was 93% (95%CI; 80.3‐100). There were no acute grade 3‐

5 radiotherapy toxicities. There was one grade 3 late toxicity; a

stricture requiring feeding tube. No treatment related deaths

occurred.

Conclusions: Combined modality treatment with radiotherapy for

early stage PG‐DLBCL has excellent local control and survival out-

comes. In our small series PG‐DLBCL appeared to have superior

outcomes when compared to nodal DLBCL. Treatment is well toler-

ated with modest toxicity.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Extranodal

non‐Hodgkin lymphoma, Radiation Therapy

No conﬂicts of interests pertinent to the abstract.

350 |DIFFUSE LARGE B‐CELL LYMPHOMA IN ELDERLY

PATIENTS: OUTCOME IN REAL‐WORLD CLINICAL PRACTICE

B. Marques

, S. Duarte

, C. Afonso

, C. B. Lima

, D. Neves

, A. C. Lai

M. J. Julião

, A. Roque

, L. Ruzickova

, J.P. Carda

, M. Gomes

, A.

Cipriano

, A. I. Espadana

Centro Hospitalar e Universitário de Coimbra, Clinical Hematology

Department, Coimbra, Portugal,

Centro Hospitalar e Universitário de

Coimbra, Pathology Department, Coimbra, Portugal

SUPPLEMENT ABSTRACTS

437

Introduction: Diffuse large B cell lymphoma (DLBCL) is the most

common type of non‐Hodgkin lymphoma, and is more frequent in

elderly patients (pts), which have a worse prognosis than younger

patients. This is partially explained by the existence of comorbidities

and poor performance and functional status, limiting the use of R‐

CHOP as standard therapy. We aim to evaluate the impact of clin-

ical factors in therapeutic decisions and outcomes of elderly DLBCL

pts.

Methods: Retrospective analysis of >65 years pts with DLBCL, NOS

diagnosed in our center, between 2010‐2019. Pts were divided in 2

groups: 66–79 years (elderly‐E) and ≥80 years (very elderly‐VE). Pts

baseline characteristics, treatment and outcomes were collected.

Multivariate analyses were performed by stepwise variable selection,

with P<0.1 in univariate analysis.

Results: We identiﬁed 193 pts, 127 (65.8%) in E group, and 66

(34.2%) in VE group. At diagnosis, 21.0% of E group and 68.8% of VE

group had an ECOG‐PS≥2 (P<0.001). The VE pts presented more

frequently with high risk IPI (22.2% vs 47.3%, P=0.006), as well as

NCCN‐IPI and GELTAMO‐IPI. Proportions of pts with bulky mass

(P=0.911) and germinal center B‐cell subtype (P=0.81) were similar

in both groups.

E group received curative intent immunochemotherapy (ICT) in

92.1% (n =117) and VE in only 53% (n =35) (P<0.001). Elderly pts

were more likely to receive rituximab (87.2 % vs 41.3%, P<0.001) and

anthracycline (88.2% vs 47.0%; P<0.001) than VE. Among pts with

curative intent ICT, complete response rate was 57.5% vs 25.8%

(P<0.001). Palliative therapy in E (n =10) and VE (n =30) groups

included steroids (70.0% vs 70.0%), PEP‐C regimen (0% vs 6.7%) or no

treatment at all (30.0% vs 23.3%). There were 14 pts (11.1%) in E group

and 6 pts (9.3%) in VE group who received consolidation radiotherapy

(RT). One patient (1.5%) in VE group received palliative RT.

The median follow‐up was 7.1(5.3‐7.2) years. Elderly pts pre-

sented a signiﬁcantly better PFS than VE (3.3[0.7‐5.9] vs 0.8[0.6‐0.9]

years, HR 1.9 [CI95% 1.4‐2.8], P<0.001) and OS (4.5[1.6‐7.4] vs 0.9

[0.4‐1.3] years, HR 2.2 [CI95% 1.5‐3.2], P<0.001).

Overall, pts who received curative intent ICT presented a

longer OS (3.3 vs 0.4 years, HR 3.3 [CI95% 2.2‐5.0], P<0.001) and

PFS (2.3 vs 0.2 years, HR 3.0 [CI95% 2.0‐4.3], P<0.001). In multi-

variate analysis, female sex, stage I/II, absence of bulky mass and

treatment with rituximab predicted a favorable prognosis in E group

for PFS and OS. In VE pts, only female sex, ECOG‐PS 0‐1 and

treatment with rituximab remained independent favorable prog-

nostic factors.

Conclusions: In our cohort, VE pts presented with a poorer functional

status and a higher risk disease, when compared to E pts. PFS and OS

was improved with curative intent ICT in both age groups, indicating

that this treatment should be considered, even in the oldest patients,

after a careful evaluation of ECOG‐PS and comorbidities.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Chemo-

therapy, Immunotherapy

No conﬂicts of interests pertinent to the abstract.

351 |RITUXIMAB AND NONPEGYLATED LIPOSOMIAL

DOXORUBINCIN (R‐NPLD) TREATMENT IN PATIENTS 80 YEARS

OF AGE OR OLDER AFFECTED BY DIFFUSE LARGE B CELL

LYMPHOMA (DLBCL): A 2020 UPDATE AND IMPLICATIONS OF

CLINICAL AND PATHOLOGICAL FACTORS

C. Cantò

, E. Pennese

, M. Di Nicola

, G. Ricciuti

, F. Restuccia

, S.

Luciani

, F. Angrilli

Civic Hospital “Santo Spirito” Pescara, Center for the Diagnosis and

Treatment of Lymphomas, Department of Oncology and Hematology,

Pescara, Italy,

“G. d'Annunzio” University, Department of Medical, Oral

and Biotechnological Sciences, Chieti, Italy,

IRCCS “Casa Sollievo della

Sofferenza”, Department of Hematology, San Giovanni Rotondo (FG), Italy

Background: In 2018 we report a rituximab plus nonpegylated lip-

osomial doxorubicin (R‐NPLD) combination for patients 80 years or

older with diffuse B cell lymphoma (DLBCL) or grade 3 b follicular

lymphoma. The overall 3‐year survival, cause‐speciﬁc survival and

progression‐free survival rates were 46%, 55%, and 44%, respectively.

According to these results, R‐NPLD has become the new standard

treatment in patients ≥80 years old with aggressive B lymphoma, in our

institution. To better investigate the prognostic role of clinical and

pathological factors, we analyzed the same immunochemotherapy

combination in a larger cohort of patient 80 years or older with DLBCL.

Methods: We retrospectively and prospectively analyzed data of pa-

tients 80 years or older with untreated histologically‐proven CD20‐

positive DLBCL, Ann Arbor stage I to IV from our institution. Pa-

tients received a combination treatment with rituximab plus non-

pegylated liposomial doxorubicin. The regimen consisted of R 375 mg/

sqm and NPLD 50 mg/sqm administered intravenously on cycle day 1,

plus prednisone 50 mg orally on days 1 to 5, every 21 days for 6

courses.

Results: Between May 2010 and April 2019, we enrolled 50 patients

(median age 84, range 80‐96, M/F:27/23). The median follow‐up time

was 28 months (range 10‐104). The overall 3‐years survival, cause‐

speciﬁc survival, and disease free survival rates were 49.9±7.6%,

55.5±7.9%, and 48.5±7.8%, respectively. Treatment was well toler-

ated with only mild toxicities, without treatment related hospitaliza-

tion or toxic deaths. Patients achieving EFS12 and EFS18 had an

overall3‐years survival of 66±13.0% and 67.9±7.0%, respectively.

Conclusion: Our results conﬁrm that, in patients 80 years or older

with DLBCL, R‐NPDL is very effective and safe combination. Among

prognostic factors, only the elevated LDH (≥1.25 upper limit)

strongly correlates with overall survival and risk of relapse, in uni-

variate (p =0.001, p =0.003) and multivariate (p =0.002, p =0.005)

analysis, respectively. In patients who achieved EFS18 the probability

to survive 24 and 36 months is of 90.5 and 67.9%, respectively. This

analysis suggests that EFS18 will be useful in patient counseling and

should be considered as a robust end point for future studies ofnewly

diagnosed very elderly DLBCL patients.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

438

SUPPLEMENT ABSTRACTS

HODGKIN LYMPHOMA

352 |BRENTUXIMAB VEDOTIN CONSOLIDATION AFTER

AUTOLOGOUS STEM CELLS TRANSPLANTATION FOR HODGKIN

LYMPHOMA: A REAL‐LIFE EXPERIENCE BY FONDAZIONE

ITALIANA LINFOMI (FIL)

F. Massaro

, V. Pavone

, P. M. Stefani

, B. Botto

, A. Pulsoni

, C.

Patti

, M. Cantonetti

, A. Visentin

, P. R. Scalzulli

, A. Rossi

, S.

Galimberti

, M. Cimminiello

, G. Gini

, M. Musso

, M. Sorio

, A.

Arcari

, V. R. Zilioli

, A. Bari

, D. Mannina

, A. Fabbri

, G.

Pietrantuono

, O. Annibali

, A. Tafuri

, E. Prete

, A. Mulè

, E.

Barbolini

, L. Marcheselli

, S. Luminari

, F. Merli

PhD Program in Clinical and Experimental Medicine, University

of Modena and Reggio Emilia, Modena, Italy,

Department of

Hematology and Bone Marrow Transplant, Hospital Card. G. Panico,

Tricase, Italy,

Hematology Unit, General Hospital Ca' Foncello,

Treviso, Italy

Division of Hematology, Città della Salute e della Scienza

Hospital and University, Turin, Italy,

Hematology Unit, Department of

Translational and Precision Medicine, Sapienza University, Rome, Italy,

Division of Hematology, Azienda Villa Soﬁa‐Cervello, Palermo, Italy,

Unit of Lymphoproliferative Disorders, Policlinico Tor Vergata,

Rome, Italy

Hematology and Clinical Immunology Unit, Department of

Medicine (DIMED), University of Padua, Padua, Italy,

Department of

Hematology, Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy,

Hematology, Azienda Socio Sanitaria Territoriale Papa Giovanni XXIII,

Bergamo, Italy,

Division of Hematology, Department of Clinical and

Experimental Medicine, University of Pisa, Pisa, Italy,

Hematology, San

Carlo Hospital, Potenza, Italy,

Division of Hematology, Azienda

Ospedaliera Universitaria Ospedali Riuniti, Ancona, Italy,

Department

of Oncology, Hematology and BMT Unit, Casa di Cura La Maddalena,

Palermo, Italy,

Department of Clinical and Experimental Medicine,

Hematology and Bone Marrow Transplant Unit, Verona, Italy,

Hema-

tology Unit, Ospedale Guglielmo da Saliceto, Piacenza, Italy,

Division

of Hematology, ASST Grande Ospedale Metropolitano Niguarda, Milan,

Italy,

Dipartimento di Scienze Mediche e Chirurgiche Materno‐Infantili

e dell'Adulto, Università di Modena e Reggio Emilia, Modena, Italy

Unit

of Haematology, Azienda Ospedaliera Papardo, Messina, Italy,

Hema-

tology Unit, Azienda Ospedaliero‐Universitaria Senese, Siena, Italy,

Hematology and Stem Cell Transplantation Unit, IRCCS Centro di

Riferimento Oncologico della Basilicata, Rionero in Vulture, Italy,

Unit

of Haematology and Stem Cell Transplantation, Campus Bio‐Medico

University, Rome, Italy,

University Hospital Sant'Andrea, Sapienza,

University of Rome, Rome, Italy,

Gruppo Amici dell'Ematologia GRADE,

Onlus Foundation, Reggio Emilia, Italy,

Fondazione Italiana Linfomi,

Onlus, Modena, Italy,

Hematology Unit, Azienda Unità Sanitaria Locale

‐IRCCS, Reggio Emilia, Italy

Introduction: Classical Hodgkin lymphoma (cHL) is a highly curable

cancer, but up to 30% of patients have a refractory or relapsed (R/R)

disease. Salvage therapy followed by consolidation with autologous

hematopoietic stem cell transplantation (ASCT) can save only

approximately half of R/R patients.

In the phase 3 AETHERA trial, cHL patients at high‐risk of pro-

gression or relapse who received Brentuximab Vedotin (BV) as

consolidation therapy after ASCT showed a reduced risk of pro-

gression compared to a placebo group, with 5‐year progression‐free

survival (PFS) rates of 59% and 41%, respectively. We report here

the results of a real‐life study on 105 cHL patients treated with BV as

consolidation after ASCT.

Methods: This retrospective study included R/R cHL patients from

15 Italian centers treated between April 2011 and August 2020.

Patients were eligible if they had received at least 2 cycles of BV

after ASCT, independently from prior treatments. The primary aim

was PFS and OS assessment and its comparison to data already re-

ported in literature.

Results: We included 105 patients, with a median follow‐up of 20

months (range 2‐108). Patients received a median of 2 lines of

treatment before ASCT. The 51% (54 patients) of total population

received BV also immediately before ASCT. PET‐CT evaluation

before and after ASCT reported a Deauville score (DS) 1‐3 in

72 (75%) and 68 (78%) patients, respectively. Considering pre ASCT

high‐risk features (refractory disease, CR <12 months, extranodal

disease at relapse), 30 (29%) patients presented at least 2 factors.

The median number of BV consolidation cycles was 10. A com-

plete schedule of sixteen cycles of BV was administered to 57 (54%)

subjects in the whole population (60% of patients who received both

consolidation and pre ASCT and 43% of subjects treated with BV

post‐ASCT only). Main causes for treatment interruption were:

adverse events (AEs; 15; 33%), disease progression (13; 28%),

consolidation with allo‐SCT (8; 17%). Among the grade 3‐4 AEs

leading to treatment interruption, we reported 8 peripheral neu-

ropathies, 4 infections, 2 infusion reactions, 1 liver toxicity.

Concerning efﬁcacy data, the 3‐year PFS and OS were 62%

(95% CI: 49‐72) and 86% (95% CI: 73‐93), respectively (Figure 1).

In univariate analysis the only feature signiﬁcantly associated with

both reduced PFS and OS was a DS 4‐5 before ASCT (HR 3.81; 95% CI:

1.80‐8.09; p <0.001). Interestingly the administration of BV pre ASCT

was not associated with different risk of progression PFS (HR 0.87,

95% CI 0.44‐1.73, p =0.965) or death (HR 0.71, 95% CI 0.20‐2.49,

p=0.594).

Conclusions: BV treatment as post ASCT consolidation is conﬁrmed by

our real‐life study as an effective and safe option for R/R cHL patients

in line with the results of the AETHERA trial. The use of BV also before

ASCT did not negatively impact on its safety and efﬁcacy, and likely

allowed to offer to a higher number of patients the option of ASCT.

Keywords: Hodgkin lymphoma, Immunotherapy

Conﬂicts of interests pertinent to the abstract

A. Pulsoni

Consultant or advisory role: Takeda

M. Cantonetti

Consultant or advisory role: Takeda

SUPPLEMENT ABSTRACTS

439

Educational grants: Takeda

A. Visentin

Educational grants: Takeda

A. Rossi

Consultant or advisory role: Takeda

V. R. Zilioli

Consultant or advisory role: Takeda

A. Fabbri

Consultant or advisory role: Takeda

Educational grants: Takeda

F. Merli

Consultant or advisory role: Takeda

353 |BRENTUXIMAB VEDOTIN PLUS BENDAMUSTINE AS

FIRST SALVAGE TREATMENT IN PATIENTS WITH RELAPSED OR

REFRACTORY HODGKIN LYMPHOMA. EXPERIENCE AT

DONOSTIA UNIVERSITY HOSPITAL, SPAIN

B. Mendibil Esquisabel

, L. Ondarra Segurola

, A. Altuna Mongelos

J. Iriondo Alzola

, A. Zumalde Murua

, A. Arambarri Oyarzabal

N. F. Rois Pego

, I. Zeberio Etxetxipia

Hospital Universitario Donostia, Hematology Department, Donostia,

Spain

Introduction: Hodgkin lymphoma (HL) is a highly curable disease

with frontline treatment. Nonetheless, up to 30% of the patients are

primarily refractory or experience relapse after upfront chemo-

therapy. Salvage chemotherapy followed by autologous stem cell

transplantation (ASCT) is the standard of care in these patients.

Achieving complete response (CR) prior to ASCT is one of the

strongest predictors of favorable outcome after ASCT. Although

Brentuximab vedotin (BV) can only be used in patients with relapsed

or refractory HL who have received at least two prior salvage

treatments, studies have shown that the combination of BV and

bendamustine as ﬁrst salvage therapy is highly active in this patient

population, increasing the number of patients who achieve CR prior

to ASCT

. We hereby report the results of patients treated with BV

and bendamustine as ﬁrst salvage regimen (out of label) in our

center.

Methods: We conducted a retrospective analysis of 12 patients un-

der 60 years with relapsed or refractory HL who received BV and

bedamustine as ﬁrst salvage regimen.

Results: Patient characteristics are shown in Table 1. A total of 12

patients (6 primary refractory and 6 relapsed) received 3 or 4 cycles

of BV plus bendamustine as ﬁrst salvage regimen. Overall, 11/12

patients achieved a response and 1 patient who had been primary

refractory progressed during treatment. Nine patients obtained a CR,

including all relapsed patients and three of the primary refractory,

and two achieved partial response (PR), both being primary re-

fractory. All patients who obtained a response underwent ASCT and

100% of the patients achieved CR at day +100 after ASCT. Three

FIGURE 1 Kaplan‐Meier plots showing OS and PFS

440

SUPPLEMENT ABSTRACTS

of the twelve patients presented high risk characteristics for

relapse after ASCT and received maintenance BV. With a median

follow‐up time of 23 months from ASCT 3 patients have relapsed at

4, 8 and 22 months, none of them having received maintenance

therapy.

No serious adverse events were observed. A total of 7 patients

experienced infusion‐related reactions (IRRs) despite premedication

with corticosteroids and antihistamine drugs. IRRs were mild‐

moderate and occurred during the second cycle. The most common

symptoms were fever, dyspnea, nausea and rash. One patient

experienced grade 2 neutropenia and received prophylactic growth

factor support.

Conclusions:

1. The combination of BV and bendamustine has shown beneﬁt

as ﬁrst salvage regimen for relapsed and refractory HL, improving the

overall response rate and CR rate prior to ASCT to 91% and 75%

respectively in our experience.

2. The treatment was well tolerated. Adverse events were mostly

IRRs of mild‐moderate intensity. In order to control IRRs a desensi-

tization protocol was applied to all patients from the second cycle on.

No patient had to discontinue therapy due to adverse events.

TABLE 1Patient characteristics and results

Appendix

SUPPLEMENT ABSTRACTS

441

REFERENCES

1. LaCasce AS, et al. Brentuximab vedotin plus bendamustine: a

highly active ﬁrst salvage regimen for relapsed or refractory

Hodgkin lymphoma. Blood. 2018 Jul 5;132(1):40‐48.

Keywords: Hodgkin lymphoma, Combination Therapies

No conﬂicts of interests pertinent to the abstract.

354 |BRENTUXIMAB VEDOTIN FOR TREATMENT IN PATIENTS

WITH RELAPSED/REFRACTORY CLASSICAL HODGKIN

LYMPHOMA IN A REAL WORLD SETTING: CLINICAL OUTCOMES

AND IMPACT ON QUALITY OF LIFE

T. Ionova

, M. Andrievskikh

, A. Amdiev

, E. Baryakh

, V. Chang

, A.

Endakova

, N. Fadeeva

, G. Husainova

, V. Ivanov

, K. Kaplanov

, O.

Kaverina

, M. Kiseleva

, T. Klitochenko

, V. Kurakin

, O.

Larionova

, D. Lazareva

, K. Lepik

, I. Lysenko

, V.

Melnichenko

, N. Mikhailova

, R. Minullina

, O. Mironov

, E.

Misyurina

, N. Mochkin

, T. Nikitina

, Y. Osipov

, T. Petrova

, N.

Porﬁrieva

, O. Rukavitsyn

, R. Saﬁn

, A. Samoylova

, T.

Shelekhova

, D. Sherstnev

, P. Simashova

, E. Smirnova

, N.

Trenina

, E. Vasiliev

, E. Volodicheva

Multinational Center for Quality of Life Research, Department of

Hematology, Saint Petersburg, Russian Federation,

Chelyabinsk Regional

Clinical Center of Oncology and Nuclear Medicine, Hematology,

Chelyabinsk, Russian Federation,

V.M. Efetov Crimean Republican

Oncology Center, Hematology, Simferopol, Russian Federation,

City

Clinical Hospital №52, Hematology, Moscow, Russian Federation,

Tambov Regional Oncological Clinical Center, Hematological, Tambov,

Russian Federation,

Kirov Research Institute of Hematology and Blood

Transfusion under the Federal Medical Biological Agency, Hematological,

Kirov, Russian Federation,

Chelyabinsk Regional Clinical Center of

Oncology and Nuclear Medicine, Hematology, Saint Petersburg, Russian

Federation,

Republican Clinical Oncology Center of the Ministry of

Health of the Republic of Tatarstan, Hematological, Kazan, Russian

Federation,

Almazov National Medical Research Centre, Department of

Anesthesiology and Intensive Care Medicine, Saint Petersburg, Russian

Federation,

S.P. Botkin City Clinical Hospital, Hematological

Department, Moscow, Russian Federation,

Altai Regional Oncology

Center, Hematological Department, Barnaul, Russian Federation,

Volgograd Regional Clinical Oncological Center, Hematological

Department, Volgograd, Russian Federation,

Clinical Oncological

Center, Hematological Department, Omsk, Russian Federation,

Primorskiy Regional Oncologic Center, Hematological Department,

Vladivostok, Russian Federation,

RM Gorbacheva Research Institute of

Pediatric Oncology, Hematology and Transplantation, Pavlov University,

Chemotherapy and Bone Marrow Transplantation Department, Saint

Petersburg, Russian Federation,

National Medical Research Centre for

Oncology of the Ministry of Health of Russia, Department of

Oncohematology, Rostov‐on‐Don, Russian Federation,

N.I. Pirogov Na-

tional Medical Surgical Center, Department of Hematology and Chemo-

therapy with a room of Bone Marrow and Hematopoietic Stem Cells

Transplantation, Moscow, Russian Federation,

Almazov National Med-

ical Research Centre, Department of Oncohematology, Saint Petersburg,

Russian Federation,

N.N. Burdenko Main Military Clinical Hospital,

Hematological Center, Moscow, Russian Federation,

Clinic of Profes-

sional Pathology and Hematology named after V.Ya. Shustov, Saratov

State Medical University, Hematology, Saratov, Russian Federation,

Regional Clinical Hospital, Hematological Department, Krasnoyarsk,

Russian Federation,

Tula Regional Clinical Hospital, Hematological

Department, Tula, Russian Federation

One of the approved treatment options for pts with relapsed/re-

fractory classical Hodgkin Lymphoma (rr‐cHL) is brentuximab vedo-

tin (BV). Comprehensive evaluation of treatment effects of

brentuximab vedotin (BV) in pts with rr‐cHL, including clinical and

patient‐reported outcomes, is worthwhile. We aimed to evaluate

clinical outcomes and quality of life (QoL) in rr‐cHL pts receiving BV

in the real world setting.

Pts with rr‐cHL receiving BV 1.8 mg/kg q3w were included in the

multicenter real‐world study. Treatment response was assessed us-

ing RECIST criteria v. 1.1. The Kaplan‐Meier method was used to

estimate progression free survival (PFS). For assessment of adverse

events (AEs) NCI CTCAE v. 4.0 was used. For QoL assessment pts

ﬁlled out RAND SF‐36. Generalized Estimating Equations (GEE) and

McNemar's test were applied.

In the total 70 pts with rr‐cHL were enrolled in the study. The

analysis was carried out in two groups. Group 1 included 62 pts

who were ineligible for ASCT or relapsed after ASCT – median age

– 31 yrs, 53% males; 50% had IV stage; 66% – B‐symptoms; 18% –

ECOG 2‐3; 19% pts in Group 1 relapsed after ASCT. Group 2

consisted of 8 pts who received BV following ASCT – median age –

28 yrs, 63% males; 25% had IV stage; 13% – B‐symptoms; no pts

with ECOG 2‐3. All the pts received ≥2 previous treatment lines

(2‐10). Before BV treatment start 60% pts in Group 1 and 38% pts

in Group 2 had severe/critical QoL impairment. In Group 1 at

median follow‐up of 8 mo, 68% pts achieved/maintained objective

response (OR). Among them 40% pts achieved complete remission

(CR) and 28% – partial remission (PR). The median PFS in Group 1

was 10.6 mo (95% CI: 7.4‐12.9). In Group 2 at median follow‐up of

15 mo 7/8 pts achieved/maintained OR; among them 6 pts had CR

and one – PR. The median PFS in Group 2 was not reached. The

most common AEs were peripheral neuropathy and fatigue (55%

and 36%). AEs of grades III‐IV in Group 1 were reported in one

patient (1.6%), SAE – one case (sepsis, respiratory insufﬁciency due

to agranulocytosis), in Group 2 – in 2 pts; one case of SAE in Group

2 was registered (abdominal syndrome). In Group 1 during BV

treatment QoL improved by all SF‐36 scales (GEE, p <.01) with the

most pronounced increase of role physical functioning (∆=60) and

role emotional functioning (∆=44) as compared to baseline. Propor-

tion of pts with severe/critical QoL impairment signiﬁcantly

decreased during BV treatment (p <.05). In Group 2 QoL stabili-

zation or improvement was obtained during BV treatment; number

of pts with severe/critical QoL impairment slightly diminished as

compared to baseline.

442

SUPPLEMENT ABSTRACTS

The results support the notable efﬁcacy and good tolerability of

BV for treatment of rr‐HL in the real‐life setting. Treatment with BV

has positive impact on QoL in rr‐cHL pts. Improvement of QoL is a

valuable outcome of BV treatment and supports patient‐

centeredness approach in treatment decision‐making in this pa-

tient's population.

The research was funded by: IIR funded by Takeda

Keywords: Hodgkin lymphoma, Immunotherapy

Conﬂicts of interests pertinent to the abstract

T. Ionova

Other remuneration: Principle investigator of the IIR

355 |EVALUATION OF RESPONSE TO THE FIRST‐LINE

THERAPY OF PATIENTS WITH CLASSICAL HODGKIN

LYMPHOMA BASED ON INTERIM PET‐CT AT A PRIVATE

BRAZILIAN INSTITUTION

J. M. De Almeida

, A. Alves

Instituto Hemomed de Oncologia e Hematologia, Clinical Department,

São Paulo, Brazil,

Instituto Hemomed de Oncologia e Hematologia,

Presidency, São Paulo, Brazil

Introduction: In Brazil the ﬁrst‐line therapy for Hodgkin lymphoma is

generally performed with ABVD protocol, being treatment strategies

guided by PET‐CT not commonly adopted. Epidemiological and

treatment data on the disease is scarce, coming mostly from public

health services.

Methods: Retrospective study of patients with classical Hodgkin

lymphoma, who were diagnosed and treated from 01/01/2016 to 12/

31/2020. Were included all patients registered on electronic health

records of the Hemomed Institute of Oncology and Hematology with

ICD.10 C81, above 18 years and with at least 3 years of follow up.

They were classiﬁed according to Clinical Stratiﬁcation Criteria

considering Ann Arbor Staging and Laboratory Markers. Then ana-

lysed the rates of Complete Response at the end of the ﬁrst‐line

treatment with AVBD protocol +/‐radiotherapy, considering strati-

ﬁcation at diagnosis and response rates on the occasion of interim

PET.

Results: 46 patients were eligible, being 69.5% female. The mean age

of diagnosis was 32 years old. Regarding staging, 43.4% presented

stage I or II and 56.6% stage III or IV. 32.6% of patients presented

bulky mass, 13% in stages I or II. 65% presented Advanced Stage

Disease according to the criteria by the European Organization for

Research and Treatment of Cancer (EORTC). PET‐CT was performed

for initial staging in 80% and interim PET in 87% of patients. Among

patients with Early Stage Disease, 80% presented a complete

response (CR) in interim PET. Of the 28 patients of the Advanced

Stage Disease Group, 19 (68%) had a CR in interim PET and 9 (32%)

had a Partial Response (PR). Analysis of response to treatment

according to classiﬁcation and interim PET shows that in the Early

Stage Groups all patients had CR in PET by the end of treatment,

including those who had obtained PR in interim PET previously.

However, in the Advanced Stage Group, of the 19 (68%) patients that

had CR in interim PET, only 16 (57%) maintained CR at the end of the

treatment. Out of the 9 (32%) patients with PR in interim PET, 5

(18%) did not have CR at the end of the treatment, taking into

consideration the fact that the data was not available for 1 of the

patients.

Conclusion: Patients from Early Stage Groups, being favorable or

unfavorable, obtained a high percentage of CR with the use of ABVD

+/‐radiotherapy. However, Advanced Group patients that continued

with ABVD, especially those with positive interim PET, presented a

high rate of refractory or early relapse. Such ﬁnding reinforces the

prognostic value of positive interim PET in Advanced Group patients

and the need to intensify therapy in these patients. This strategy is of

key importance to reducing the need for rescue therapy.

Keywords: PET‐CT, Hodgkin lymphoma

Conﬂicts of interests pertinent to the abstract

J. M. De Almeida

Other remuneration: Aché Laboratórios Farmacêuticos S. A.

356 |THE ROLE OF END OF TREATMENT PET CT EVALUATED

BY DEAUVILLE FIVE‐POINT SCALE AS PROGNOSTIC ROLE IN

HODGKIN LYMPHOMA

G. M. Assanto

, R. Agrippino

, G. Lapietra

, M. L. De Luca

, A.

Chiaravalloti

, G. Annechini

, G. M. D'Elia

, G. Ciotti

, M. Martelli

, A.

Pulsoni

Haematology, Department of Translational and Precision Medicine,

Sapienza University of Rome, Rome, Italy,

Nuclear medicine, Department

of Biomedicine and Prevention, Nuclear Medicine, University Tor Vergata,

Rome, Italy, Rome, Italy

Introduction: Positron Emission Tomography/Computed Tomogra-

phy (PET CT) is crucial in staging and response assessment in

Hodgkin lymphoma (HL). While early tumor response is assessed

through an interim PET (I‐PET) by Deauville 5‐point scale (DS), end of

treatment (EoT) PET CT is used to evaluate absence or presence of

metabolic disease, lacking scientiﬁc evidence of a correlation be-

tween speciﬁc DS and prognosis. Our study aimed to analyse EoT

PET CT response to demonstrate a possible prognostic correlation

between DS and patient prognosis in terms of Relapse Free Survival

(RFS) and Overall Survival (OS).

Methods: We conducted a monocentric retrospective study in pa-

tients with Classic HL, consecutively treated with ABVD between

2007 and 2018 with at least 1 year of follow‐up and with favourable

I‐PET (DS1‐3). EoT and I‐PET CT images were collected and sub-

mitted to blind central revision and DS assessment. Different values

of DS at EoT PET CT were compared in terms of PFS. Survival

SUPPLEMENT ABSTRACTS

443

analysis was performed by Kaplan‐Meier curves and Log‐Rank test.

Statistical signiﬁcance was considered for values of p <0.05.

Results: PET CT images of 78 patients were centrally reviewed. All

patients are currently alive (OS=100%). After a median follow‐up of

60 months (range 17‐139) 17 patients (21%) had disease recurrence,

with RFS of 60% at 104 months (median not reached). The median

time to relapse was 8 months (range 3‐39). Splenic involvement,

anemia (Hb<12 gr/dl) and bulky disease at diagnosis increased risk of

relapse. Patients with EoT DS1 (56 cases) showed 83% RFS at 100

months (median not reached). Median RFS worsened for higher DS:

77 months for DS2 (12 cases), 2 months for DS3 (2 cases), 26 months

for DS 4 (3 cases), and 14 months for DS5 (p <0.001) [Shown in

ﬁgure]. There was a longer time frame to relapse in DS1/2 compared

to DS≥3 (median time of 34 and 4 months respectively). Comparing I‐

PET DS with EoT PET DS, higher RFS was observed in stable or

reduced metabolic activity, unlike worsened DS indicates increased

risk of relapse (p <0.001).

Conclusion: Our study suggests that a systematic evaluation of EoT

PET according to DS allows more accurate identiﬁcation of patients

with an unsatisfactory metabolic response and a better prognostic

stratiﬁcation. The joint evaluation of the I‐PET and EoT PET data

shows a higher risk of recurrence in case of increased DS. In patients

ending treatment with a persistently signiﬁcant FDG uptake some

form of consolidation could be investigated.

Figure Relapse free survival (RFS) according to End of Treatment

(EoT) score of Deauville 5‐point scale (DS): patients with a DS 1 show

a better prognosis. A DS progressively higher is indicative of residual

metabolic uptake and is associated with an increased risk of relapse

(p <0.001).

Keywords: Diagnostic and Prognostic Biomarkers, PET‐CT, Hodgkin

lymphoma

No conﬂicts of interests pertinent to the abstract.

357 |A PROSPECTIVE UKRAINIAN MULTICENTRE COHORT

STUDY OF PET VALUE PROGNOSTIC ROLE IN PRIMARY

PATIENTS WITH HODGKIN LYMPHOMA (HL): HIGHLIGHTING

CLINICAL RELEVANCE

O. Novosad

, I. Pastushenko

, O. Gorbach

, V. Kozlov

, L.

Mykhalska

, O. Kindrakevych

, M. Novikov

, O. Karnabeda

, O.

Oliinichenko

, Y. Kmetyuk

, O. Karpova

, A. Ashykhmin

, T. Rudiuk

I. Kriachok

National Cancer Institute, Oncohematology, Kyiv, Ukraine,

Regional

Clinical Hospital, Hematology, Odessa, Ukraine,

Clinical Hospital

“Feofaniya”, Center of Hematology, Kyiv, Ukraine,

Oncology Hospital

“LISOD”, Radiology, Kyiv, Ukraine,

Hospital of Oncology “Innovacia”,

Hematology, Kyiv, Ukraine,

Center of Nuclear Medicine, Radiology, Kyiv,

Ukraine,

Clinical Hospital “Feofaniya”, Radiology, Kyiv, Ukraine,

National Cancer Institute, Radiology, Kyiv, Ukraine,

Bogomolets

National Medical University, Oncology, Kyiv, Ukraine

Introduction: Using PET in a metabolic response assessment has

provided an opportunity of individualized treatment for patients with

newly diagnosed HL. Current risk assessment systems cannot pro-

vide a reliable prediction of failure in an individual treatment. The

primary endpoints were to assess a correlation between PET ﬁndings,

regimens, IPS and survival, as well as whether there is a difference

when patient treatment is intensiﬁed or not intensiﬁed depending on

PET ﬁndings.

Methods: This sub‐study was conducted in a scope of a broader

prospective multicenter study on the predictive value of PET

response assessment (205 patients with HL, up to 69 years of age).

444

SUPPLEMENT ABSTRACTS

Patients received a treatment with ABVD or BEACOPP‐14/esc or

“switched‐regimens” (ABVD+BEACOPP‐esc/14, BEACOPP‐esc/14

+ABVD). PET imaging was performed using Deauville criteria pro-

tocol for response assessment. According to IPS, 155 (75.6%) pa-

tients had ≤2 and 50 (24.4%) patients had 3‐4 risk factors (low and

high risk groups, respectively, p <0.05).

Results: The ORR was 91.7%. The ORR for patients with stage I‐II

was 95.5% (106/111) vs 87.2% (82/94) for patients with stage III‐IV.

In total, the disease progression occurred in 48.3% (15/31) of

interim PET‐positive (PET2+) patients and 11.5% (20/174) of interim

PET‐negative (PET2‐) patients (p <0.05). We conﬁrmed a signiﬁcant

difference between the 5‐year event free survival (EFS) rate for

patients with PET2+vs PET2‐. Thus, PET2‐patients with early and

advanced stages had an EFS level much higher than the PET2+pa-

tient group (87% and 85% vs 55% and 15% respectively, Log‐rank

test, p =0.0001).

Using ABVD and “switched‐regimens” for I‐II‐stage patients did

not reveal a signiﬁcant correlation to the 5‐year EFS. However, a more

intensiﬁed treatment for patients with stage III‐IV of HL showed a

signiﬁcantly better level of EFS compared to the ABVD regimen, and it

did not depend on PET2 status (Log‐rank test, p =0.0003). Moreover,

intermediate‐risk group cohort (2B stage) tends to improve the EFS

levels when the regimens was given BEACOPP‐esc/14+ABVD vs

ABVD+BEACOPP‐esc/14 vs ABVD (80% vs 60% vs 55% respectively,

p=0.064) and was independent of PET2 status.

Patients in low risk group were associated with signiﬁcant

improvement in 5‐year EFS compared with patients in high risk group

(83% vs 50%, respectively, p =0.01). The estimated 5‐year EFS was

85% in PET2‐vs 45% with PET2+in low risk group (p <0.0001). The

same result was found in high risk group patients (82% vs 50%,

p<0.0001).

Conclusion: Our results of PET2 and PET3 remains very important in

the early detection of high‐risk patients treated with standard

chemotherapy regimens, who are unlikely to show a complete

response to therapy, which may justify intensiﬁcation of treatment at

that time, contributing to improved survival. According to IPS,

stratiﬁcation still has an important role affecting the patients' with

HL survival.

EA – previously submitted to regional or national meetings (up to

1000 attendees) and EHA 2021.

Keywords: PET‐CT, Hodgkin lymphoma, Chemotherapy

No conﬂicts of interests pertinent to the abstract.

358 |DO WE STICK TO INTERNATIONAL GUIDELINES WHEN

TREATING EARLY STAGE HODGKIN LYMPHOMA?

A. C. Oliver

, V. Irigoin

, V. Bradvica

, V. Lema

, A. I. Landoni

, C.

Ambrosoni

, P. Kollar

, A. Remedi

, S. Damiano

, P. Muxi

Hospital Británico, CASMU, Departamento de Hematología, Montevideo,

Uruguay,

COSEM, CASMU, Departamento de Hematología, Montevideo,

Uruguay,

Hospital Militar, Departamento de Hematología, Montevideo,

Uruguay,

Hospital Maciel, Departamento de Hematología, Montevideo,

Uruguay,

Medica Uruguaya, Departamento de Hematología, Montevideo,

Uruguay,

Instituto Nacional del Cancer, Departamento de Hematología,

Montevideo, Uruguay,

Hospital Británico, Departamento de

Hematología, Montevideo, Uruguay

Introduction: Early stage Hodgkin Lymphoma (HL) presents in >60%

of newly diagnosed patients, and cure is achieved in >90%. Since

2003 combined modality with chemo (CT) and radiotherapy (RT) has

been the standard of care and reduction of long term toxicities is a

priority. Several trials have proved the efﬁcacy of reducing CT cycles

and dose of RT. The German HD‐10 trial showed that 2 ABVD plus

ISRT 20 Gy in favorable early stage HL was as effective as more

intensive treatment and is a widely recommended strategy. In 2015

RAPID trial showed that 3 ABVD when PET negative could be suf-

ﬁcient. In stage I‐II unfavorable HL, 4 ABVD plus ISRT or 6 ABVD are

classical options. The aim of our study is to analyze to what extent we

have adopted the management proposed by the guidelines in daily

practice in our country.

Materials: HL patients between 2011‐2020 from the Uruguayan HL

Registry. 153 patients included, 68 stage I‐II. Prognostic: German

Study Group factors.

Results: 68 patients: 27 male. Median age: 35.5 years (15‐83).

Nodular sclerosis: 52 (76.5%). Stage I: 11 (16.2%), stage II 57 (83.8%).

Favorable 37 (54.4%), unfavorable 31 (45.6%).

Stages I‐II FAVORABLE: 36 patients. Male 14 (48.9%). Median

age 41 (15‐83). Stage I: 8 (22.2%), stage II: 28 (77.8%). 33 received

ABVD (91.7%), 2 AVD (5.6%), 1 supportive care.

Median number of cycles: 4 (2‐6); stage I: 2(2‐6), stage II: 4 (2‐6).

• 2 ABVD +RT: 5 patients, 14.3 %

• 3 ABVD: 3; 2 +RT, 8.6%

• 4 ABVD: 12, 34.3 %; 4 +RT, 11.43%

• 6 ABVD: 13; 1 +RT, 37.2 %

• 2 AVD +RT: 1, 2.8 %

• 4 AVD: 1, 2.8 %

End of treatment‐PET (EOT‐PET): 29 (80.6%). Complete

response (CR): 33 (91.7%). Median follow‐up 40 months (0.7‐134), 1

relapse (2.8%), 2 deaths: 1 due to relapse, 1 due to untreated pro-

gression. 1 year Progression free survival (PFS) 97 %, 2 year Overall

Survival (OS) 92.9%.

Stages I‐II UNFAVORABLE: 32 patients. Male 13 (30.6%). Me-

dian age: 30 (15‐75). Stage I: 3 (9.4%), II: 29 (90.6%). 31 received

ABVD (96.9%), 1 AVD (3.1%), 1 still on treatment. Median number of

cycles: 6 (1‐8); stage I: 6 (4‐6), stage II: 6 (1‐8).

• 1 ABVD: 2 patients, 6,4 %

• 4 ABVD: 11, 35.5 %; 3 plus RT, 9.7 %

• 6 ABVD: 16; 4 plus RT, 51.7 %

• 8 ABVD: 1, 3.2 %

• 6 AVD: 1, 3.2 %

EOT‐PET: 22 (68.8%). CR: 21 (65.6%). Median follow‐up 57.4

months (0.7‐142). 3 year PFS was 87%. 5 year OS 90.6%.

SUPPLEMENT ABSTRACTS

445

Conclusions: From 2011 and according to the HD10 study, RAPID

trial and NCCN guidelines stage I‐II favorable HL should be treated

with ABVD x 2 +ISRT 20 Gy, 3 ABVD or 4 ABVD +ISRT. Only 34.3

% of our cohort of patients were treated according to any of these

recommendations. Regarding stage I‐II unfavorable HL, guidelines

suggest ABVD x 4 +ISRT or 6 ABVD +/‐ISRT; 61.4% of our patients

were treated according to them.

This reﬂects that changes in traditional management of patients

take time to become standard in daily practice after evidence is

presented. Even though PFS and OS are comparable to international

data, we should improve our approach in order to reduce toxicity

without compromising results.

Keywords: Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

359 |ANALYSIS OF SPERMATOGENESIS BEFORE AND AFTER

TREATMENT IN PATIENTS WITH HODGKIN LYMPHOMA

F. Gaudio

, P. Masciopinto

, C. Nardelli

, M. P. Vacca

, F. E. Laddaga

E. Cicinelli

, P. Musto

AOUC Policlinico, Unit of Hematology and Stem Cell Transplantation,

Bari, Italy,

AOUC Policlinico, Unit of Second Unit of Obstetrics and

Gynecology, Bari, Italy,

AOUC Policlinico, Clinical Pathology Unit, Bari,

Italy

The outcome of patients with Hodgkin's lymphoma (HL) has

improved signiﬁcantly in recent years, which is why attention has

increasingly been focused on the well‐being of these young patients.

Consistent with this, the study aimed to analyze the inﬂuence of HL

and its treatment on spermatogenic status of young males and detect

new predictors inﬂuencing probability of parenthood.

In this retrospective study, spermatogenesis of 42 male patients

with HL (median age 25 years) treated at University Hospital of Bari

(Italy) between 2008 and 2016 was studied. Spermiogram was avail-

able in all patients at diagnosis, while in 24 spermiogram was also

available after therapy. All patients underwent ABVD as ﬁrst‐line

chemotherapy, of which 6 patients with relapsed/refractory disease

underwent autologous haematopoietic stem cell transplantation

(ASCT).

Results: At diagnosis, we found that 15 (35%) patients had normo-

zoospermia, 19 (45%) oligozoospermia, and 8 (19%) only other dys-

permic conditions.

Analyzing prognostic factors, we found that number of sperma-

tozoa at diagnosis was reduced in stage 3‐4 (1.91x10e7 vs.

4.97x10e7; p =0.001); motility and vitality were reduced in stage 3‐4

(motility: 29% vs. 43%, p =0.015; vitality: 49% vs 71%, p =0.011)

and in presence of B symptoms (motility: 30% vs. 44%, p =0.016;

viability: 53% vs. 69%, p =0.05); abnormal forms were increased in

patients with high ESR (84% vs. 75%, p =0.017) and albumin <4 (87%

vs. 75%, p =0.002).

After treatment, 17% of patients had azoospermia. Of the 9

normozoospermic patients at diagnosis, 5 (55%) maintained this

condition after treatment, while 4 (45%) developed azoospermia

associated with other types of dyspermia (all undergoing ASCT). Of

11 patients with oligozoospermia at diagnosis treated with ﬁrst line

alone, 6 (54%) experienced normospermia after treatment. Further-

more, in patients who received only ﬁrst‐line treatment, the mean

sperm concentration increased after treatment (5.53x10e7 at diag-

nosis vs. 4.02x10e7 after treatment).

We also found that 4/6 patients undergoing ASCT had azoo-

spermia in the post‐treatment control. On the other hand, 2 of

the 3 patients, who had oligozoospermia before treatment,

became normozoospermic after therapy. Furthermore, we found

that patients undergoing ASCT were associated with severe

impairment of fertility in terms of sperm motility (74% at diagnosis

vs. 22% after ASCT; p =0.025). In patients who did not undergo

ASCT we found a statistically signiﬁcant improvement in fertility in

terms of motility (35% at diagnosis vs. 50% after treatment;

p=0.009)

Conclusions: In this study, we found that ASCT induced infertility in

the majority of male patients with HL and that ﬁrst line treatment

could improve the fertility status caused by the disease. Further

studies are needed in a larger case series to investigate risk factors

for impaired fertility at diagnosis and after treatment

Keywords: Late Effects in Lymphoma Survivors

No conﬂicts of interests pertinent to the abstract.

PERIPHERAL T/NK CELL LYMPHOMAS

360 |BRENTUXIMAB VEDOTIN PLUS CHP AS FIRST‐LINE

TREATMENT IN CD30 +PERIPHERAL T‐CELL LYMPHOMAS: REAL

LIFE EXPERIENCE FROM A SINGLE INSTITUTION IN SPAIN

E. Domingo Domenech

, J. M. Sancho Cia

, E. Gonzalez Barca

N. Kelleher

, M. Rodriguez‐Luaces

, J. Rovira

, S. Verdesoto

M. Encuentra

, D. Blazevic

, A. Oliveira

, L. Escoda

, J. M. Ribera

A. Sureda

Institut Català d'Oncologia, Hospital Duran i Reynals, Institut

d'Investigació Biomèdica de Bellvitge (IDIBELL), Universitat de Barcelona,

Hematology, L'Hospitalet de Llobregat, Barcelona, Spain,

Institut Català

d'Oncologia, Hospital Germans Trias i Pujol, Institut de Recerca contra

la Leucèmia Josep Carreras, Universitat Autònoma de Barcelona,

Badalona, Barcelona,Spain., Hematology, Badalona, Spain,

Institut

Català d'Oncologia, Institut d'Investigació Biomèdica de Girona

(IDIBGI), Universitat de Girona, Hematology, Girona, Spain,

Institut

Català d'Oncologia, Hospital de Tortosa Verge de la Cinta, Hematology,

Tortosa, Spain,

Institut Català d'Oncologia, Hospital Universitari

Joan XXIII, Hematology, Tarragona, Spain,

Institut Català

d'Oncologia, Hospital Moises Broggi, Hematology, Sant Joan d'Espí,

Barcelona., Spain

446

SUPPLEMENT ABSTRACTS

Introduction: Peripheral T‐cell lymphomas (PTCL) are a heteroge-

neous group of lymphomas classically treated with CHOP or CHOP‐

like regimens, with poor outcomes. CD30 is universally expressed

and is pathognomonic in systemic anaplastic large cell lymphoma

(sALCL), with variable expression among non‐sALCL PTCL subtypes

(40‐60%). Recent data of frontline treatment with Brentuximab

Vedotin (BV), an anti‐CD30 monoclonal antibody, plus CHOP has

demonstrated signiﬁcant improvement in survival (ECHELON‐2

clinical trial), becoming the new standard of care for sALCL in Europe.

Patient and methods: From February 2019 to December 2020, 20

patients with de novo newly diagnosed CD30+PTCL have been

treated with the combination of BV‐CHP with primary prophylaxis

with G‐CSF, in the centers of the Catalan Institute of Oncology in

Spain. Survival curves were plotted by the Kaplan‐Meier method.

Results: Clinical characteristics at diagnosis are shown in the table.

Of interest, 5 of the 10 ALK negative ALCL patients were diagnosed

of breast implant associated ALCL (BIA‐ALCL) with extracapsular

involvement. The number of cycles administrated was 102, with a

median of 6 cycles per patient (range 1‐6). At the time of this report,

2 patients were still on treatment. Seven cycles (7%) were delayed (3

due to infection, 2 due to neutropenia grade 2, and 2 due to causes

not related with chemotherapy). An adverse event was reported in

45 (44%) cycles, being the most frequent peripheral neuropathy in

14, nausea/vomiting in 9 and anemia in 8; all of them grade 1‐2.

Treatment was discontinued after 1 cycle in 1 patient due to pro-

gression. Of the 18 evaluable patients, the overall response rate

(ORR) was of 83%, with 72% complete responses and 11% partial

responses. Consolidative autologous stem cell transplant (ASCT) was

performed in 5 patients. With a median follow‐up of 14 months

(limits: 1‐24), 1‐year progression‐free survival (PFS) and overall

survival (OS) was 68.2% (95% CI 44.6‐91.7) and 82.2% (95% CI 63.9‐

100), respectively.

Conclusions: Brentuximab Vedotin plus CHP is an effective regimen

for CD30 positive PTCL, with a high rate of response. This combi-

nation presents a manageable safety proﬁle, with the majority of

patients completing the planned treatment. The incidence and

severity of side effects are low, being peripheral neuropathy and

neutropenia the most frequent.

Keywords: Aggressive T‐cell non‐Hodgkin lymphoma, Combination

Therapies

No conﬂicts of interests pertinent to the abstract.

361 |RISK MODEL FOR CENTRAL NERVOUS SYSTEM DISEASE

IN T‐LYMPHOBLASTIC LYMPHOMA: A SINGLE‐CENTER

EXPERIENCE

X. Feng

, Y. Gao

, L. Li

, X. Li

, Z. Sun

, J. Wu

, X. Wang

, X. Fu

, L.

Zhang

, M. Zhang

The First Afﬁliated Hospital of Zhengzhou University, Department of

Oncology, Zhengzhou, China

Introduction: T‐lymphoblastic Lymphoma (T‐LBL) represents for 2‐

4% of adult lymphoma, characterized by mediastinal mass and

young male predominance. Central nervous system (CNS) prophy-

laxis has become an integral part of all current acute lymphoblastic

leukemia treatment protocols. However, there is no consensus on

CNS prophylaxis in T‐LBL. Therefore, we retrospectively summa-

rized prognosis of CNS disease and risk factors for CNS relapse in

T‐LBL.

TABLE 1Clinical characteristics at diagnosis

SUPPLEMENT ABSTRACTS

447

Methods: We retrospectively collected 81 newly diagnosed T‐LBL

patients in our single institution from January 2015 to December

2019. CNS disease was diagnosed via contrasted magnetic resonance

imaging at the presence of typical symptoms, and/or detection of

lymphoma cells in the cerebrospinal ﬂuid by either morphology or

immunophenotypes.

Fisher's exact test or chi‐square test was used for discrete var-

iables, and t test or Mann‐Whitney test was used for continuous

variables. Kaplan‐Meier method was used to predict risk factors for

CNS disease. Statistical analyses were performed using SPSS, and p <

0.05 was considered to be statistically signiﬁcant.

Results: Median age of the cohort was 25 years. 69 patients (85.2%)

were Ann Arbor III‐IV stage, 18 patients (22.2%) presented with B

symptom. 11 cases (22.4%) were identiﬁed as Early T‐cell precursor

(ETP) LBL.

During a median follow‐up of 20.5 months (3.0‐60.5 months),

there were totally 10 patients (12.3%) diagnosed CNS disease. 3

cases (3.7%) were diagnosed at the initiation of treatment, 7 pa-

tients experienced CNS relapse during the treatment course. Me-

dian time to CNS disease was 7 months (0‐20.0 months), and

median survival after CNS disease was 1.5 months (0.5‐16.0

months).

The 2‐year progression‐free survival (PFS) of the entire cohort

was 39.63%, 2‐year overall survival (OS) was 54.11%. Median PFS

and OS of patients with CNS disease was much less than those

without CNS disease.

The univariable analysis showed elevated LDH, Ki67≥80%, ETP

phenotype and more than 2 sites of extra‐nodal involvement pre-

dicted CNS disease. While the multivariable cox analysis revealed

that Ki67≥80% (HR =165.3, P=0.011), ETP phenotype (HR =31.1,

P=0.024) and more than 2 sites of extra‐nodal involvement

(HR =21.4, P=0.037) were independent risk factors for CNS disease

in T‐LBL.

Subsequently, these three independent risk factors were com-

bined (1 point for each risk factor) to create a risk‐stratiﬁcation

model called CNS‐LBL. Two CNS‐LBL subgroups were identiﬁed as

having low (0‐1 risk factor), and high (2‐3 risk factors). 38 cases

(69.1%) were low‐risk subgroup, 19 patients (30.9%) belonged to

high‐risk group. 2‐year CNS relapse rates were 5.4% and 88.7%,

respectively (P=0.000), resulting in a 16.5 folds higher risk for CNS

relapse in the high‐vs low‐risk groups.

Conclusions: CNS disease displays a dismal prognosis in T‐LBL, and

our retrospective study generated a risk model to identify high‐risk

populations.

The research was funded by: The study was supported by the

National Natural Science Foundation of China (Grant No.82000203)

Keywords: Diagnostic and Prognostic Biomarkers, Aggressive T‐cell

non‐Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

362 |DERIVATION AND VALIDATION OF A NOVEL LIPID‐

COVERED PROGNOSTIC SCORING SYSTEM FOR NEWLY

DIAGNOSED MATURE T AND NK CELL LYMPHOMAS

T. Lu

, X. Fang

, Y. Jiang

, J. Liu

, Y. Cai

, S. Hu

, M. Ding

, X. Wang

X. Zhou

Shandong Provincial Hospital Afﬁliated to Shandong University,

Department of Hematology, Jinan, China

Introduction: Abnormal lipid metabolism was reported to inﬂu-

ence the initiation and progression of malignancies. Nevertheless,

the signiﬁcance of dyslipidemia in mature T and natural killer

(NK) cell lymphomas remains unexplored. This study aimed to

investigate the association between serum lipid metabolism level

and clinical outcomes, and further develop a lipid‐covered prog-

nostic model for newly‐diagnosed mature T and NK cell lym-

phoma patients.

Methods: A total of 173 patients (1/2006‐6/2020) were enrolled in

the study, of which 115 constituted the training cohort while 58

composed the validation cohort. Univariate and multivariate Cox

regression analysis were performed to detect independent predictors

and generate a novel prognostic model, whose performance was

evaluated in the training and validation cohort from multiple aspects,

including Harrell's C‐index, calibration curves, Kaplan‐Meier survival

analysis, time‐dependent receiver operating characteristic curves,

Brier score and decision curve analysis.

Results: Univariate analysis veriﬁed declined cholesterol (TC), high‐

density lipoprotein cholesterol (HDL‐C) and increased triglycerides

(TG) were associated with inferior survival outcomes in mature T and

NK lymphomas. Multivariate analysis revealed extranodal involved

sites≥2 (HR: 2.505; P=0.031), β2‐MG≥3mg/L (HR: 3.964; P=0.005)

and TC<3.58mmol/L (HR: 3.320; P=0.000) were three independent

predictors. Thus, a novel prognostic model, EnBC score, was estab-

lished with these three variables. EnBC score possessed satisfactory

discrimination and calibration since the Harrell's C‐index for OS

prediction was 0.840 (95% CI, 0.810‐0.870) and the calibration

curves showed an optimal agreement between the prediction and

actual observation. The prognostic signiﬁcance of the novel model

was veriﬁed in Kaplan‐Meier survival analysis by dividing patients

into four risk groups with different long‐term survival (median OS, NA

vs. NA vs. 14.0 vs. 4.0 months, P<0.0001; median PFS, 84.0 vs. 19.0

vs. 8.0 vs. 1.5 months, P<0.0001) (Figure 1). Compared with the

present International Prognostic Index (IPI), EnBC score presented

better discrimination (time‐dependent AUCs of OS, EnBC 0.883‐

0.963 vs. IPI 0.785‐0.873, P<0.001) and better calibration (Brier

score of OS, EnBC 0.146‐0.271 vs. IPI 0.154‐0.296, P=0.023).

Moreover, the decision curve analysis showed that EnBC score pro-

vided larger net beneﬁts than IPI in the entire patient population.

Conclusions: Serum lipid metabolism level was ﬁrstly demonstrated

signiﬁcantly prognosis‐related in mature T and NK cell lymphomas.

448

SUPPLEMENT ABSTRACTS

Furthermore, a novel lipid‐covered prognostic scoring system was

established and performed well in predicting the prognoses of

treatment‐naïve mature T and NK cell lymphoma patients.

Keywords: Diagnostic and Prognostic Biomarkers, Aggressive T‐cell

non‐Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

363 |EXTRANODAL NK/T‐CELL LYMPHOMA, NASAL TYPE

AND HIV INFECTION: A MEXICAN DESCRIPTIVE SERIES

R. Castillo‐llanos

, A. F. Ramírez‐Ibarguen

, C. Lome‐Maldonado

Instituto Nacional de Cancerología, Hematología, Mexico City, Mexico,

Instituto Nacional de Cancerología, Hematopathology, Mexico City,

Mexico

SUPPLEMENT ABSTRACTS

449

Introduction: Extranodal NK/T‐cell lymphoma, nasal type (ENKL), is

a rare and aggressive lymphoma, characterized by an extranodal

involvement with midfacial destruction, primarily in nasal and

paranasal structures, and is highly associated with Epstein‐Barr

virus (EBV). It is an uncommon lymphoma with higher incidence

in Asian and South American countries. Infection with Human

Immunodeﬁciency Virus (HIV) is a known risk factor for develop-

ment of non‐Hodgkin lymphoma. In most of the cases are B‐cell

types, with a smaller percentage being T‐cell lymphomas. There

are few case reports of this association, and no large series was

found.

Methods: It was made a research on a data base of all lymphomas

and HIV from 2010 to 2020, ENKL pathology diagnose and HIV

diagnosed conﬁrmed by a 4th gen ELISA cases were selected. Clinical

data from the electronic medical records were extracted. The main

objective was to describe the clinical characteristics of our HIV and

ENKL series.

Results: Of a total of 258 cases of lymphoma plus HIV, 7 (2.7%)

belonged to newly diagnostic of ENKL. The median age at lymphoma

diagnosis was 39 years (21 – 57) and all were male. Four of the patients

knew his HIV status at diagnosis, the other three were diagnosticated

at the same time of the lymphoma diagnosis. Only three of the patients

known as a HIV positive were on High Activity Antiretroviral Therapy

(HAART) at the diagnosis, the rest start HAART after diagnosis. At the

diagnosis of ENKL only 2 patients had a CD4 count >150 cells, and

only 2 had undetectable viral load; in one patient this was not deter-

mined at diagnosis. In all patients the ENKL was nasal type, only one

patient presented with a stage IV for lung involvement, 2 patients were

stage II by nodal involvement, the rest were localized at midfacial

structures. All cases were associated with EBV with EBER positive, 3

cases were positive for CD30. Regards to treatment, one patient died

before treatment because AIDS complications and six patients

received treatment with L‐asparaginase based chemotherapy (CHT)

(SMILE, LVP, GeLOx, Asp/Met/Dex). Three of patients received CHT

followed by radiotherapy (RT) with a median dose of 44 Gy (36 – 55),

these patients achieve complete response. One more present pro-

gression at central nervous system (SNC) after two cycles of Asp/Met/

Dexa protocol and went to second line but died for infective compli-

cations 11 months after diagnosis. Two patients received one or two

cycles and lost follow up before ended treatment. At the end, 3 pa-

tients still alive with mean follow up was 16 months (2 – 61).

Conclusions: ENKL is highly associated with EBV and actually we

know that HIV patients may have early exposition to EBV. A few has

been written about ENKL and HIV association, for this we describe

the clinicopathologic characteristics of our patients with these

concomitant diseases. It's important to note the rate of CD30 pre-

sent in our patients for the role of therapeutic target.

Keywords: Aggressive T‐cell non‐Hodgkin lymphoma, Extranodal

non‐Hodgkin lymphoma, Non‐Hodgkin (Pediatric, Adolescent, and

Young Adult)

No conﬂicts of interests pertinent to the abstract.

NEW DRUGS

364 |THE CD19/CD3 BISPECIFIC ANTIBODY WORK

EFFECTIVELY AS ADJUNCT WITH IBRUTINIB ON THE

TREATMENT OF B‐CELL LYMPHOMA

H. Yu

, W. Liu

, L. Mi

, S. Shu

, W. Zhang

, Z. Ying

, H. Chen

, X. Yan

W. Shen

, G. Tu

, Y. Ye

, M. Li

, D. Wang

, D. Hu

, J. Cao

, F. Qi

X. Wang

, Y. Song

, J. Zhu

Beijing Cancer Hospital, Lymphoma, Beijing, China,

Beijing Cancer

Hospital, Molecular Oncology Laboratory, Beijing, China,

ITabMed Ltd,

Preclinical Research, Shanghai, China

Introduction: Targeting CD19, a signiﬁcant surface marker of B cell

malignant cells, with anti‐CD19/CD3 bispeciﬁc antibodies became

new choices in B cell hematological malignancies. Blinatumomab as the

ﬁrst generation CD19/CD3 bispeciﬁc antibody (bsAb) was approved

for the treatment of relapsed or refractory B‐cell precursor acute

lymphoblastic leukemia. However, it also has its own limitations,

including side effects and limited efﬁcacy. Bruton tyrosine kinase

(BTK) inhibitor, ibrutinib has been shown to inhibit the IL‐2‐induced

tyrosine kinase (ITK) which expressed in T and NK cells. This inhibi-

tion could result in modulation of cytokine production by T cells. Here

we reported a newly designed CD19/CD3 bispeciﬁc antibody (A‐319)

with similar mechanism of action as blinatumomab. To further improve

the efﬁcacy and reduce side effects, we tried to combine ibrutinib with

A‐319, and explore the biological effects and the mechanisms of

enhanced efﬁcacy and reduced CRS in vitro and in vivo.

Methods: A‐319 was provided by ITabMed Ltd (Shanghai, China).

CFSE was used to evaluate the proliferation of T cells, and CD25/CD69

staining was applied to assess the activation of T cells induced by A‐

319. Supernatant from co‐cultured tumor and T cells was collected

and measured for lactate dehydrogenase (LDH) lysis to measure

cytotoxicity. Humanized mice model were built by human PBMC in-

jection into NOG mice. To build CRS model, we allowed Raji derived

tumor to grow up to 600‐800mm3, at which point we dosed mice with

high dose A‐319(20μg/kg). Tube formation assay using HUVEC cells

was performed to explore the effects of ibrutinib on angiogenesis.

Results: A‐319 effectively induced CD19‐positive tumor cell lysis.

The regression analysis presented a signiﬁcant correlation between

CD19 expression of target cells and the sensitivity to A‐319

(p <0.01). Exposure T cells to A‐319 enhanced T cell proliferation

and promoted T cell activation (CD25+CD69+T cells) signiﬁcantly.

Moreover, A‐319 produced remarkable anti‐tumor effects in vivo

though companied by CRS. Interestingly, the addition of ibrutinib

inhibited A‐319 induced CRS and improve the anti‐tumor efﬁcacy in

vitro and in vivo. Experiments exploring the mechanisms of the

combined effects showed that ibrutinib could inhibit the inﬁltration

of M2 (CD163, Arg‐1) macrophage and normalize vascular (CD34+α‐

SMA+) in vivo compared with A‐319. The in vitro results also pre-

sented that ibrutinib could polarize macrophage into M1 (CD80,

CD86, iNOS). Tube formation assay on HUVEC cells proved that

ibrutinib could inhibit macrophage mediated tube formation.

450

SUPPLEMENT ABSTRACTS

Conclusions: We found that A‐319 could greatly activate T cells and

promote the proliferation of T cells from patients. The addition of

ibrutinib effectively enhanced the anti‐tumor effects and lower the

CRS of A‐319 possibly through the regulation of macrophage medi-

ated angiogenesis.

Keywords: Tumor Biology and Heterogeneity, Combination Thera-

pies, Immunotherapy

No conﬂicts of interests pertinent to the abstract.

365 |PRECLINICAL EVALUATION OF EPCORITAMAB

COMBINED WITH STANDARD OF CARE AGENTS FOR THE

TREATMENT OF B‐CELL LYMPHOMAS

C. W. Chiu

, I. H. Hiemstra

, W. ten Hagen

, R. Snijdewint‐Nkairi

, B.

de Jong

, P. Garrido Castro

, C. Kweekel

, R. S. Oliveri

, B. Elliot

, M.

Wielgos‐Bonvallet

, E. Szafer‐Glusman

, D. Schuurhuis

, J. Blaedel

T. Ahmadi

, E. Breij

, A.K. Sasser

, M. Jure‐Kunkel

Genmab US, Inc., Translational Research, Plainsboro, New Jersey, USA,

Genmab B.V., Translational Research, Utrecht, Netherlands,

Genmab A/

S, Oncology, Copenhagen, Denmark,

Genmab US, Inc., Oncology,

Plainsboro, USA,

Abbvie, Translational Oncology, Sunnyvale, California,

USA

Introduction: B‐cell non‐Hodgkin lymphoma (B‐NHL) has a high

unmet medical need for efﬁcacious, well tolerated, off‐the‐shelf

therapies that can combine with standard of care (SOC) regimens.

Epcoritamab (DuoBody

‐CD3�CD20) is a subcutaneously (SC)

administered bispeciﬁc antibody with a manageable safety proﬁle

and promising preliminary anti‐tumor activity in aggressive and

indolent B‐NHL. Epcoritamab simultaneously binds to CD3 on T

cells and CD20 on tumor cells to induce potent T‐cell‐mediated

killing. SOC treatments for B‐NHL include GEMOX (gemcitabine

and oxaliplatin), or rituximab in combination with chemotherapy

regimens such as CHOP (cyclophosphamide [CPA], doxorubicin

[DOX], vincristine [VINC] and prednisone [PRD]), bendamustine, or

immunomodulatory agents such as lenalidomide (LEN). These ther-

apies have mechanisms of action distinct from epcoritamab and

mostly non‐overlapping adverse event proﬁles. Preclinical studies

showed that epcoritamab's anti‐tumor activity is retained in the

presence of a rituximab analog. Here, we present in vitro data of

epcoritamab in combination with components of current SOC

therapies for B‐NHL.

SUPPLEMENT ABSTRACTS

451

Methods: Healthy donor T cells (effector cells) and/or CD20‐

expressing B‐NHL tumor cell lines were pre‐treated with SOC com-

ponents (LEN, CPA, DOX, VINC or PRD) to evaluate the impact on T

cells and target cells. Pre‐treated T and B‐NHL cells were incubated

with the same SOC in the presence of epcoritamab, and T‐cell‐

mediated cytotoxicity and associated T‐cell activation were

assessed by ﬂow cytometry. Bendamustine or GEMOX was added

with epcoritamab during the T‐cell activation and cytotoxicity assay

to assess potential antagonizing effects.

Results: LEN enhanced T‐cell activation induced by CD3 crosslinking

with immobilized anti‐CD3 or epcoritamab, resulting in higher potency

of these T cells to exert epcoritamab‐induced cytotoxicity of CD20‐

expressing tumor cells. T cells pre‐treated with individual CHOP

components could mediate epcoritamab‐induced cytotoxicity. Finally,

bendamustine and GEMOX did not antagonize T‐cell activation and

increased T‐cell‐mediated cytotoxicity induced by epcoritamab.

Conclusion: These preclinical data support further exploration of

combination of epcoritamab with these SOC agents in clinical trials.

Given the promising single‐agent activity of SC epcoritamab, and

distinct safety proﬁle, these combinations may lead to further

deepening of treatment responses and transform long‐term outlook

for B‐NHL patients. Epcoritamab in combination with these SOC

agents is currently being evaluated in a clinical trial (NCT04663347).

EA – previously submitted to AACR 2021.

The research was funded by: Genmab US, Inc.

Keywords: Combination Therapies

Conﬂicts of interests pertinent to the abstract

C. W Chiu

Employment or leadership position: Genmab US, Inc.

Stock ownership: Genmab US, Inc.

I. H Hiemstra

Employment or leadership position: Genmab

Stock ownership: Genmab

W. ten Hagen

Employment or leadership position: Genmab

R. Snijdewint‐Nkairi

Employment or leadership position: Genmab

B. de Jong

Employment or leadership position: Genmab

P. Garrido Castro

Employment or leadership position: Genmab

C. Kweekel

Employment or leadership position: Genmab

R. S Oliveri

Employment or leadership position: Genmab

Stock ownership: Genmab

B. Elliot

Employment or leadership position: Genmab

Stock ownership: Genmab

M. Wielgos‐Bonvallet

Employment or leadership position: Genmab

E. Szafer‐Glusman

Employment or leadership position: Abbvie

D. Schuurhuis

Employment or leadership position: Genmab

J. Blaedel

Employment or leadership position: Genmab

Stock ownership: Genmab

T. Ahmadi

Employment or leadership position: Genmab

Stock ownership: Genmab

E. Breij

Employment or leadership position: Genmab

Stock ownership: Genmab

A K. Sasser

Employment or leadership position: Genmab

Stock ownership: Genmab

M. Jure‐Kunkel

Employment or leadership position: Genmab

Stock ownership: Genmab

366 |IRONOMYCIN KILLS DIFFUSE LARGE B‐CELL

LYMPHOMA CELLS BY TARGETING CELLULAR IRON

HOMEOSTASIS

J. Devin

, T. Cañeque

, Y.‐L. Lin

, L. Mondoulet

, J.‐L. Veyrune

, M.

Abouladzé

, E. Garcia de Paco

, O. Gadacha

, G. Cartron

, P. Pasero

C. Bret

, R. Rodriguez

, J. Moreaux

IGH‐CNRS‐UMR 9002, Institute of Human Genetics, Montpellier, France,

CNRS UMR 3666. INSERM U1143, Chemical Biology of Cancer Labo-

ratory, Paris, France,

Sideros, Paris, France,

CHU Montpellier, UTC,

Montpellier, France,

CHU Montpellier, Clinical Hematology, Montpellier,

France,

CHU Montpellier, Biological Hematology, Montpellier, France

Introduction: Diffuse large B cell lymphoma (DLBCL) is the

most common hematological malignancy. Although more than half of

these patients may achieve long‐term remission, the majority of

the remaining patients succumb to DLBCL. Abnormal iron homeostasis

452

SUPPLEMENT ABSTRACTS

is implicated in carcinogenesis and in the progression of many tumors.

Thus, we searched for a putative altered iron homeostasis in DLBCL

that could be exploited to develop novel therapeutic strategies.

Method: In this study, we analyzed the iron metabolism gene expres-

sion proﬁle (GEP) of large cohorts of DLBCL patients. We also analyzed

the effect and mechanism of action of ironomycin, a synthetic deriva-

tive of salinomycin that presents an antineoplastic effect on breast

cancer stem cells by accumulating and sequestering iron in lysosomes.

We used a large panel of 16 DLBCL cell lines and primary samples from

patients collected after patients' written informed consent in accor-

dance with the Declaration of Helsinki and institutional research board

approval.

In vivo validation was investigated using the A20 syngeneic

immunocompetent model of murine lymphoma. We also investigated

the effect of ironomycin combined with conventional drugs

commonly used in DLBCL. Ironomycin‐associated hematopoietic

toxicity was investigated in vitro using hematopoietic progenitor

colony‐forming unit assays.

Results: We established the Iron Score (IS), a gene expression‐based

risk score (validated in 4 independent cohorts of patients) allowing to

identify DLBCL patients with a poor outcome and that could beneﬁt

from a suitable targeted therapy. We analyzed the therapeutic

beneﬁt of ironomycin, a new promising lysosomal iron targeting small

molecule. Ironomycin inhibited DLBCL cell proliferation in a panel of

16 DLBCL cell lines, at nanomolar concentrations compared to

typical iron chelators. Ironomycin induced signiﬁcant cell growth in-

hibition, ferroptosis and autophagy in DLBCL cells. Upon ironomycin

treatment, we also identiﬁed DNA double strand break accumulation,

delayed progression of replication forks and a signiﬁcant increase of

RPA2 phosphorylation that is a marker of replication stress. Impor-

tantly, ironomycin signiﬁcantly reduced the median number of viable

primary DLBCL cells of patients without major toxicity for non‐tumor

cells from the microenvironment and presented a low toxicity on

hematopoietic progenitors compared to conventional treatments.

Interestingly, we identiﬁed a signiﬁcant synergistic effect by

combining ironomycin with Doxorubicin, BCL2, BTK or Syk inhibitors.

Conclusion: Altogether, these data demonstrate that a subgroup of

high‐risk DLBCL patients can be identiﬁed with the IS, which can

potentially beneﬁt from targeting iron homeostasis.

The research was funded by: The R.R. research group is funded by

the European Research Council (ERC) under the European Union's

Horizon 2020 research and innovation programme (grant agreement

No [647973]), the Fondation Charles Defforey‐Institut de France and

Ligue Contre le Cancer (Equipe Labellisée). S.D. is funded by the

Agence Nationale de la Recherche. We thank the CNRS, INSERM and

Institut Curie for generous funding. The J.M research group was

supported by grants from INCa (Institut National du Cancer)

PLBIO18‐362 PIT‐MM and PLBIO19 FATidique, ANR (TIE‐Skip;

2017‐CE15‐0024‐01), ANR‐18‐CE15‐0010‐01 PLASMADIFF‐3D,

SIRIC Montpellier Cancer (INCa_Inserm_DGOS_12553), Labex Epi-

GenMed and Institut Universitaire de France. Work in the PP lab is

supported by grants from the Agence Nationale pour la Recherche

(ANR), Institut National du Cancer (INCa), the Ligue Nationale Contre

le Cancer (équipe labellisée) and the Fondation MSDAvenir.

Keywords: Risk Models, Chemotherapy, Aggressive B‐cell non‐

Hodgkin lymphoma

No conﬂicts of interests pertinent to the abstract.

367 |SYNERGISTIC ANTICANCER EFFECT OF COMBINED

CHIDAMIDE AND ETOPOSIDE TREATMENT AGAINST NK/T CELL

LYMPHOMA

W. Song

, C. Shi

, Y. Gao

The First Afﬁliated Hospital of Zhengzhou University, Department of

Oncology, Zhengzhou, China

Background and purpose: Natural killer (NK)/T cell lymphoma

(NKTCL) is a highly aggressive hematological malignancy without

worldwide consensus on ﬁrst‐line therapy for refractory/relapsed (RR)

patients. Chidamide is a selectively histone deacetylases (HDACs) in-

hibitor. It is suggested that combined HDACs inhibitors and DNA‐

damaging agent can enhance anticancer effect. Etoposide is a well‐

known chemotherapy agent which targets type II topoisomerases,

and has been applied to treating hematological malignancies. Thus, in

this study, we investigated the synergistic anticancer effect and po-

tential mechanisms of combined chidamide and etoposide treatment in

NKTCL.

Materials and methods: A series of in vitro and in vivo experiments

were conducted using NKTCL cell lines, YT, NKYS, and KHYG‐1. The

inhibitory effect of chidamide or etoposide alone and their combi-

nation was detected using CCK‐8 assay. Cell apoptosis and cell cycle

arrest were measured using ﬂow cytometry, morphological changes

and mitochondrial membrane potential were observed using micro-

scopy, and the expression of related proteins was examined using

western blotting. Furthermore, xenograft mouse models of NKTCL

were also established to verify the synergistic effect in vivo.

Results: We found that both chidamide and etoposide showed time‐

and dose‐dependent inhibitory effect on NKTCL cell lines, and there

remained synergism of the two drugs. Combined chidamide and

etoposide treatment signiﬁcantly promoted cell apoptosis, induced

cell cycle arrest at the G2/M phase, increased histone acetylation,

decreased mitochondrial membrane potential, and caused great DNA

damage in NKTCL cell lines. In vivo, the volume and weight of the

tumor was obviously decreased due to treatment.

Conclusions: Both chidamide and etoposide inhibited the prolifera-

tion of NKTCL cell lines and showed synergistic effect in vivo and in

vitro. Combined chidamide and etoposide treatment might cause

greater DNA damage through increasing histone acetylation, arres-

ted G2/M phase, and decreasing MMP to promote apoptosis and

inhibit proliferation.

Key Words: NK/T cell lymphoma; Chidamide; Etoposide

EA – previously submitted to regional or national meetings (up to

1000 attendees).

SUPPLEMENT ABSTRACTS

453

The research was funded by: National Science and Technology

Major Project of China (grant number 2020ZX09201‐009), National

Natural Science Foundation of China (grant number 81970184,

U1904139) and Department of Science & Technology of Henan

province (grant number 182102310114).

Keywords: Combination Therapies, Extranodal non‐Hodgkin

lymphoma

No conﬂicts of interests pertinent to the abstract.

368 |THE POTENTIAL EFFICACY AND MECHANISM OF

BENDAMUSTINE IN ENTRA‐NODAL NK/T CELL LYMPHOMA

X. Feng

, Y. Gao

, H. Li

, M. Jin

, W. Song

, Z. Li

, M. Zhang

The First Afﬁliated Hospital of Zhengzhou University, Department of

Oncology, Zhengzhou, China

Introduction: Entra‐nodal NK/T cell lymphoma (NKTCL) is an

aggressive malignant lymphoma, with a prevalence of East Asia.

Nowadays, despite of the promising improvement, there remains an

urgent need of novel agent to improve the dismal prognosis of

relapse and refractory patients.Bendamustine combines the proper-

ties of purine analogue and alkylating agent, which has been suc-

cessfully used for B lymphoid neoplasms. However, its use in T cell

lymphoma is limited, especially in NKTCL. Herein, we describe for the

ﬁrst time the potential efﬁcacy of bendamustine in NKTCL.

Methods and materials: NKTCL cell lines (NKYS, KHYG‐1, NKL) were

obtained from ExPASy. Bendamustine, chloroquine, concanamycin A

and baﬁlomycin were purchased from Selleckchem. Cell proliferation

was detected by CCK8. Cell apoptosis and cell cycle were detected

using commercial kit from Keygen Biotech. Antibodies used for west-

ern blot and immunoﬂuorescence were all from Cell Signaling Tech-

nology. Nude mice were purchased from Gempharmatech and

incubated with 2*10

KHYG‐1 cells subcutaneously, then 30mg/kg

bendamustine or PBS were injected via tail vein once the tumor

appeared.

Results: Bendamustine inhibited proliferation of NKTCL cell lines in a

time and dose‐dependent manner. Interestingly, we noticed a syn-

ergistic effect of bendamustine with either gemcitabine or etoposide.

Mechanistically, bendamustine remarkably induced apoptosis of

NKTCL, along with upregulated cleavage of PARP, caspase‐3 and

caspase‐7, while Bcl‐2 was signiﬁcantly downregulated. Moreover,

bendamustine induced G2/M cell cycle arrest, which was attributed

to increased p‐cdc2 and p‐cdc25c, while decreased cyclinB1. Subse-

quently, we observed signiﬁcant upregulated DNA damage response

(DDR) using immunoﬂuorescence of p‐γH2A.X after bendamustine

exposure. Meanwhile, DDR related proteins, including p‐ATM, p‐

ATR, p‐Chk1, p‐Chk2, p‐P53 and p‐γH2A.X, were all increased in

both time and dose‐dependent way.

To further explore the mechanism of bendamustine, we pre-

treated NKTCL cell lines with autophagy and lysosomal pathway

inhibitors, and found that suppression of bendamustine on NKTLC

cells were partially abrogated. Elaborate study showed that bend-

amustine induced autophagy‐lysosome related cell death, evidenced

by increased level of LC‐3II after bendamustine treatment and

further upregulated after baﬁlomycin treatment.

Finally, we veriﬁed effect of bendamustine on NKTCL cells in

vivo. It showed that bendamustine dramatically inhibited growth of

subcutaneous tumor, with no obvious impact on mice weight.

Conclusions: Collectively, bendamustine dramatically suppressed

NKTCL both in vitro and in vivo, most possibly caused by activated

DDR and subsequent cell apoptosis and cell cycle arrest. Autophagy

and lysosomal related cell death also contributed inhibitory effect of

bendamustine.

The research was funded by: The study was supported by the

National Natural Science Foundation of China (Grant No.82000203)

Keywords: Extranodal non‐Hodgkin lymphoma, Chemotherapy

No conﬂicts of interests pertinent to the abstract.

369 |DEEP LEARNING TO PREDICT THE EFFECT OF DRUGS IN

PATIENTS WITH LYMPHOMA BASED ON CELL LINE DATA

D. K. Chebanov

, I. N. Mikhaylova

, N. S. Tatevosova

BioAlg Corp., Research and Development, Walnut, California, USA,

NMRC N.N. Blokhin, Surgery 10, Moscow, Russian Federation,

New

York Medical College, Healthcare Organization, Valhalla, New York, USA

Patient resistance to targeted drugs, even in the presence of the

actionable biomarkers, is a signiﬁcant problem, including in the

treatment of patients with lymphoma. At the same time, data from

drug testing on cell lines can help solve this problem, since machine

learning can be used to draw an analogy between the molecular

proﬁle of a cell line and a patient's tumor.

The data were taken from the compound sensitivity data for cell‐

lines database Genomics of Drug Sensitivity in Cancer (GDSC), and

also gene expression proﬁles of the lines from Cancer Cell Line

Encyclopedia (CCLE).

First, we determined the effect of the response for cell lines in

the GDSC database: it was if the IC50 value was lower than the

maximum tested concentration.

Then we added the gene expression data from CCLE database by

merging both datasheets on cell line names. In the expression data

were selected only those most widespread genes that belong to the

cancer‐related pathways (328 genes) The resulting array consisted of

449 unique compounds and 626 cell lines, in total the datasheet had

242k rows. 94k cases has the response as effect, the rest did not

have the response. So the ﬁrst step was to assess the quality of the

binary classiﬁcation to predict the response to the drug. For this

purpose we calculated the sensitivity and speciﬁcity of the classiﬁ-

cation with receiver characteristic curves (ROC) as the area under

the curve (AUC).

454

SUPPLEMENT ABSTRACTS

The neural network had 2 hidden layers each with 200 neurons

and was created with Keras library for python 3. The learning process

consisted of 10 epochs. We selected 90 drugs (20%) that showed

ROC AUC higher than 0.8.

On the second step we predicted the response of these drugs for

patients based on their tumor gene expression data. For that purpose

we took the patients gene expression data for the same genes from

the pan‐cancer TCGA database, and selected only patients with the

lymphoid neoplasm diffuse large B‐cell lymphoma (47 patients). Then

we used the same neural network to make the prediction.

As a result, we got the probability of the response for the drug.

For half of the drugs, no patients were found to have a response

rate greater than 70%. For the remaining drugs, the proportion of

patients for whom a greater than 70% response rate was predicted

ranged from 2 to 100%.

The high quality of prediction shows that deep learning is a

powerful tool for predicting drug response. A similar approach can

also be used for groups of drugs to predict the likelihood of inhibition

of a biological mechanism or signaling pathway. The described

method can be used to predict the results of clinical trials on patients,

to extrapolate the results of cell lines drug response to patients. With

the aim of further improvement, the list of genes for prediction will

be expanded, and feature selection techniques will be tested.

Keywords: Bioinformatics; Computational and Systems Biology Ge-

nomics, Epigenomics, and Other ‐Omics, Risk Models

Conﬂicts of interests pertinent to the abstract

D. K Chebanov

Employment or leadership position: CEO

Stock ownership: Yes

370 |A PHASE 1 STUDY OF CARFILZOMIB WITH RITUXIMAB,

IFOSFAMIDE, CARBOPLATIN AND ETOPOSIDE (C‐RICE) IN

TRANSPLANT‐ELIGIBLE RELAPSED/REFRACTORY DIFFUSE

LARGE B‐CELL LYMPHOMA

P. Torka

, A. Groman

, J. Wong

, B. Baysal

, J. Nichols

, A. Kader

C. Mavis

, S. Jani Sait

, A. Block

, E. Przespolewski

, A. Mohr

I. Lund

, K. McWhite

, J. Kostrewa

, J. DeMarco

, M. Johnson

A. Darrall

, R.‐N. Thomas

, S. Sundaram

, P. Ghione

, A. Hutson

F. Hernandez‐Ilizaliturri

Roswell Park Comprehensive Cancer Center, Medicine, Buffalo, USA,

Roswell Park Comprehensive Cancer Center, Biostatistics, Buffalo, USA,

Roswell Park Comprehensive Cancer Center, Pathology, Buffalo, USA,

Roswell Park Comprehensive Cancer Center, Clinical Research Services,

Buffalo, USA,

Roswell Park Comprehensive Cancer Center, Immunology,

Buffalo, USA,

Roswell Park Comprehensive Cancer Center, Cytogenetics,

Buffalo, USA,

Roswell Park Comprehensive Cancer Center, Pharmacy,

Buffalo, USA

Background: The CORAL study highlighted the need to develop novel

salvage regimens in R/R DLBCL previously treated with R‐CHOP.

Overall response rate (ORR) to second line therapy (either RICE or

RDHAP) was 51% with a 3‐year progression free survival (PFS) of only

30%. More recently, the SCHOLAR‐1 study and the anti‐CD19 CAR‐T

cell therapy trials have highlighted the dismal outcomes of pts with

chemo‐refractory disease. We previously demonstrated that 1) the

ubiquitin‐proteasome system (UPS) plays an important role in ac-

quired rituximab‐chemotherapy (R‐C) resistance in DLBCL and that 2)

carﬁlzomib (CFZ), a selective, irreversible proteasome inhibitor,

overcomes R‐C resistance in lymphoma pre‐clinical models. We hy-

pothesized that targeting the UPS with CFZ could result in improved

outcomes with 2

line therapy and high‐dose chemotherapy with

autologous stem cell transplant (HDC‐ASCT) in R/R DLBCL pts.

Methods: We conducted a single‐center, open‐label, prospective

phase 1 study evaluating the safety, efﬁcacy, and pharmacokinetics/

pharmacodynamics (PK/PD) of CFZ in combination with R‐ICE in

HDC‐ASCT eligible R/R DLBCL pts (NCT01959698). Classic 3+3 dose

escalation schema was used to determine the MTD. Primary objec-

tives were to determine the maximum tolerated dose (MTD) and

examine the dose‐limiting toxicities (DLT) of CFZ in combination with

R‐ICE. Secondary objectives include preliminary evaluation of activity,

feasibility of successful mobilization of autologous stem cells and PK/

PD analysis. Patients received CFZ (at 6 dose levels) on days 1, 2, 8, 9,

and standard doses of rituximab‐ICE on days 3‐6. Cycles were

repeated every 21 days for a maximum of 3 cycles before HDC‐ASCT.

FIGURE 1 Improved PFS and OS was noted in pts with non‐GCB DLBCL compared to GCB DLBCL with C‐RICE therapy in the R/R setting.

SUPPLEMENT ABSTRACTS

455

Toxicity was assessed using NCI CTCAE version 4; responses were

assessed after cycles 2 and 3 using revised Cheson criteria.

Results: In the dose escalation phase, 18 pts were enrolled at 6 dose

levels with no DLTs noted. CFZ 27 mg/m

was selected as the MTD.

11 additional pts were enrolled in the dose expansion phase. Median

age was 62 (29‐75) yrs, 55% were male, 34% had germinal center B‐

cell like (GCB) subtype and 62% had non‐GCB subtype DLBCL by

Han's algorithm, data was missing for 1 pt. MYC and BCL2/BCL6

rearrangements were noted in 3 pts, an additional 2 pts had an iso-

lated MYC rearrangement. Best ORR was 62% (45% CR, 15% PR),

52% pts underwent HDC‐ASCT. Interestingly, an ORR of 100% was

observed in pts with non‐GCB DLBCL. Median PFS was 15.2 months

(5.1 mo‐not reached) and median OS was NR (10.2 mo‐NR). Pts with

non‐GCB subtype had a signiﬁcantly longer PFS (NR vs 7 mo, p =

0.02) and OS (NR vs 10.7 mo, p =0.002) than those with GCB‐

subtype (Figure 1).

Conclusion: C‐RICE is well tolerated in pts with R/R DLBCL with

toxicities comparable to RICE therapy. Improved outcomes observed

in non‐GCB DLBCL highlight the contribution of the UPS in acquiring

rituximab‐chemotherapy resistance. Our data shows that pts with

non‐GCB DLBCL beneﬁt from incorporating CFZ into second line

therapy.

The research was funded by: Amgen Inc.

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Chemo-

therapy, Combination Therapies

No conﬂicts of interests pertinent to the abstract.

371 |PRELIMINARY ACTIVITY OF DEVIMISTAT (CPI 613) IN

PATIENTS WITH RELAPSED OR REFRACTORY BURKITT

LYMPHOMA/LEUKEMIA

L. Nikolaenko

, T. S. Pardee

, R. Steiner

, J. S. Abramson

, S. Horwitz

M. Matasar

, C. Owens

, I. Rodriguez‐Rivera

, D. Straus

, S. Luther

A. Noy

City of Hope, Hematology & Hematopoietic Cell Transplantation, Duarte,

CA, USA,

Wake Forrest Baptist Health, Hematology, Winston Salem, NC,

USA,

MD Anderson Cancer Center, Lymphoma‐Myeloma, Houston, TX,

USA,

Massachusetts General Hospital Cancer Center, Medicine/Lym-

phoma, Boston, MA, USA,

Memorial Sloan Kettering Cancer Center,

Medicine/Lymphoma, New York, USA,

Rafael Pharmaceuticals, Chief

Business Ofﬁcer, Cranbury, NJ, USA

Introduction: Patients (pts) with primary refractory or relapsed

Burkitt lymphoma/leukemia (BL) or high‐grade B‐cell lymphoma with

MYC and BCL2 rearrangements (double hit, DHL) and/or BCL6 (triple

hit, THL) have a dismal prognosis. A dysregulated MYC oncogene has

downstream effects on proliferation and highly glycolytic metabolism

with tricarboxylic acid (TCA) cycle intermediates as biosynthetic

precursors. Devimistat is a non‐redox active analogue of lipoic acid, a

required cofactor for two key TCA cycle mitochondrial enzymes:

pyruvate dehydrogenase and alpha ketoglutarate dehydrogenase.

Disruption results in a shutdown of ATP and biosynthetic‐

intermediate production leading to cancer cell death. In a phase I

trial (n =26) a pt with multiply refractory BL had a partial remission

sustained over one year prior to resection. She remains in remission 7

years later. As of March 2021, 20 clinical studies in over 700 pts with

various cancers have been conducted (ongoing/completed) with

devimistat. We initiated a phase II trial to further explore efﬁcacy.

Methods: NCT03793140 is a multicenter study enrolling 17 patients

on each of two cohorts BL or DHL/THL. Pts must have had one prior

therapy or are refusing standard of care, measurable disease or iso-

lated bone marrow involvement, and must not be within 3 months of a

prior stem cell transplant. Active central nervous system (CNS)

parenchymal disease is excluded, but leptomeningeal disease is

allowed if the CSF is negative for more than 4 weeks and the main-

tenance intrathecal/intraOmmaya therapy is ongoing. Devimistat is

given by central line over 2 hours daily x 5 days for two 14‐day cycles

and then as maintenance every 21 days. Pts are evaluable for

response if they receive at least 4/5 days of the ﬁrst cycle. At least

1response in the ﬁrst 9 pts by cycle 3 was needed to expand to 17 pts

in total in that cohort.

Results: 9 pts were enrolled in the DHL/THL arm. Number of prior

therapies was 3(1‐6). No responses and only 1 stable disease resulted

in cohort closure. Thus far, 8 BL pts were enrolled. Number of prior

therapies was 3 (2‐4). 2 pts were inevaluable. 1/6 pts had a response.

This HIV+pt had a thigh mass having received 4 prior therapies. He

had a near complete metabolic remission after 3 cycles of devimistat.

PET/CT assessment: Thigh lesion baseline October 19, 2020: 10.8 x

6.5 cm SUV 24. After cycle 3: unmeasurable, SUV 2. He is currently in

cycle 7, having had a 4‐week treatment delay of cycle 5 due to CoVID

19 infection. ECOG improved from 3 to 0. Adverse events (AE): As of

March 03, 2021, no serious AEs related to study drug. 4 pts had

grade 3 events at least possibly related: 2 neutropenia, 1 thrombo-

cytopenia and 1 elevated bilirubin. 1 patient had a dose reduction for

grade 2 alanine aminotransferase increase.

Conclusions: Although our results are preliminary, the near complete

remission in this patient is promising in a disease where no viable

treatment options exist.

The research was funded by: Raphael Pharmaceuticals

Keywords: Molecular Targeted Therapies, Aggressive B‐cell non‐

Hodgkin lymphoma

Conﬂicts of interests pertinent to the abstract

456

SUPPLEMENT ABSTRACTS

L. Nikolaenko

Research funding: Rafael

T. S Pardee

Consultant or advisory role: Rafael Pharmaceuticals

Research funding: Rafael Pharmaceuticals:

R. Steiner

Research funding: Rafael

J. S Abramson

Research funding: Rafael

M. Matasar

Consultant or advisory role: Genentech Bayer Merk Juno Thera-

peutics Roche Teva Rocket Medical Seattle Genetics Daiichi Sankyo

Takeda

Stock ownership: Merk

Honoraria: Genentech Roche GlaxoSmithKline Bayer Pharmacy-

clics Janssen Seattle Genetics Immmunovaccine Technologies Takeda

Research funding: Genentech Roche GlaxoSmithKline IGM Bio-

sciences Bayer Pharmacyclis Janssen Rocket Medical Seattle Ge-

netics Immunovaccine Technologies

S. Luther

Employment or leadership position: Rafael Pharmaceuticals

A. Noy

Consultant or advisory role: Janssen, Morphosys

Honoraria: Pharmacylics

Research funding: Raphael Pharma, Pharmacyclics, NIH

372 |PRELIMINARY SAFETY DATA FROM PATIENTS (PTS)

WITH RELAPSED/REFRACTORY (R/R) B‐CELL MALIGNANCIES

TREATED WITH THE NOVEL B‐CELL LYMPHOMA 2 (BCL2)

INHIBITOR BGB‐11417

C. Y. Cheah

, E. Verner

, C. S. Tam

, J. Hilger

, Y. Gao

, J. Huang

, D.

Simpson

, S. Opat

Sir Charles Gairdner Hospital and Pathwest Laboratory Medicine,

Nedlands; Medical School, University of Western Australia, Crawley;

Linear Clinical Research, Department of Haematology, Nedlands,

Australia,

Concord Repatriation General Hospital, Concord; University of

Sydney, Haematology, Sydney, Australia,

Peter MacCallum Cancer

Centre, Melbourne; University of Melbourne, Parkville; St Vincent's

Hospital, Fitzroy; Royal Melbourne Hospital, Haematology, Parkville,

Australia,

BeiGene (Beijing) Co., Ltd., Beijing, China and BeiGene USA,

Inc., Hematology, San Mateo, California, USA,

Monash Health and

Monash University, Haematology, Clayton, Australia

Background: BCL2, a key regulator of the apoptotic pathway, is

aberrantly expressed in many hematologic malignancies. Treatment

with BCL2 inhibitor venetoclax can be limited by common mild

gastrointestinal toxicities, neutropenia, and emergence of BCL2

mutations. BGB‐11417 was developed as a potent and highly selec-

tive inhibitor of BCL2 and has shown antitumor activity superior to

venetoclax in acute lymphoblastic leukemia, mantle cell lymphoma

(MCL), and diffuse large B‐cell lymphoma (DLBCL) xenograft models.

BGB‐11417 has a favorable pharmacokinetic proﬁle with low plasma

clearance in rodents and dogs. Toxicology study results have shown a

broad safety window and excellent safety proﬁle.

Methods: BGB‐11417‐101 is a ﬁrst‐in‐human phase 1/1b study

(dose escalation and safety expansion) to determine safety, tolera-

bility, maximum tolerated dose (MTD), and recommended phase 2

dose of BGB‐11417 in pts with R/R B‐cell malignancies

(NCT04277637). For dose escalation, pts were enrolled in 1 of 5 oral

BGB‐11417 dose levels (40, 80, 160, 320, or 640 mg once daily) and

received weekly or daily dose ramp‐up; pts with non‐Hodgkin lym-

phomas (NHLs) received 2‐day ramp‐up (day 1, 25% of intended

dose; day 2, 50%) before reaching the intended daily dose (day 3+,

100%). Adverse events (AEs) were reported per CTCAE v5.0. A

Bayesian logistic regression model was used to evaluate dose‐limiting

toxicities (DLTs; during ramp‐up through day 21 at intended daily

dose) and determine the MTD. The ﬁrst dose‐escalation cohort

allowed pts with R/R follicular lymphoma (FL), marginal zone lym-

phoma (MZL), DLBCL, or transformed NHL.

Results: As of 01/01/2021, 7 pts with R/R NHL had been treated. A

cohort of R/R chronic lymphocytic leukemia (CLL) had just opened

with 2 pts treated. Only data from the NHL pts are reported (5

DLBCL, 1 FL, and 1 MZL), with median follow‐up of 2.9 mo (range,

1.7‐7.7). The 40‐mg (n =3) and 80‐mg (n =4) dose cohorts are

complete with no DLTs. The 160‐mg dose cohort is underway. AEs

occurring in >1 pt are listed in Table 1. Four pts discontinued (dis-

ease progression [n =3; 2 at 40 mg, 1 at 80 mg] or lack of efﬁcacy

[n =1 at 40 mg]), and 3 pts remain on treatment. No pt discontinued

from AEs, and no instances of laboratory or clinical tumor lysis

syndrome were observed.

Conclusion/Summary: These early phase 1 results suggest that BGB‐

11417 is tolerable in pts with R/R NHL at dose levels tested. Pre-

liminary activity in this pt population will be assessed with increased

enrollment and follow‐up. Enrollment of pts with R/R CLL is un-

derway, and decreases in lymphocyte count have been seen at the

initial ramp‐up dose of 1 mg. Evaluation of pts with MCL and Wal-

denström macroglobulinemia, and the combination of BGB‐11417

and Bruton tyrosine kinase inhibitor zanubrutinib, is planned for

future cohorts.

EA – previously submitted to EHA 2021.

SUPPLEMENT ABSTRACTS

457

The research was funded by: BeiGene (Beijing) Co., Ltd., Beijing,

China and BeiGene USA, Inc., San Mateo, CA, USA

Keywords: Aggressive B‐cell non‐Hodgkin lymphoma, Molecular

Targeted Therapies

Conﬂicts of interests pertinent to the abstract

C. Y. Cheah

Consultant or advisory role: Roche, Janssen, MSD, Gilead, Ascentage

Pharma, Acerta, Loxo Oncology, TG Therapeutics

Honoraria: Roche, Janssen, MSD, Gilead, Ascentage Pharma, Acerta,

Loxo Oncology, TG Therapeutics

Research funding: Celgene, Roche, AbbVie

Educational grants: Roche

E. Verner

Research funding: Janssen‐Cilag

C. S. Tam

Honoraria: Janssen, Abbvie, Beigene

Research funding: Janssen and Abbvie

J. Hilger

Employment or leadership position: BeiGene

Stock ownership: BeiGene

Educational grants: BeiGene

Y. Gao

Employment or leadership position: BeiGene, Ltd.

Stock ownership: BeiGene, Ltd.

Educational grants: BeiGene, Ltd.

J. Huang

Employment or leadership position: BeiGene

Stock ownership: BeiGene

Educational grants: BeiGene

Other remuneration: Patents, Royalties, other Intellectual Property:

BeiGene

D. Simpson

Employment or leadership position: BeiGene

Consultant or advisory role: Abbvie, Roche Janssen

Stock ownership: BeiGene

Research funding: Roche, MSD, Acerta, Pharmacyclics, Sanoﬁ, GCK,

Jannsen, Abbvie, Celgene, Amgen

S. Opat

Consultant or advisory role: Roche, Janssen, Abbvie, Celgene,

Takeda, Merck, Gilead, Mundipharma, AstraZeneca, CSL

Honoraria: Roche, Janssen, Abbvie, Celgene, Takeda, Merck, Gilead,

AstraZeneca

Research funding: Beigene, Roche, Janssen, Abbvie, Takeda, Merck,

Gilead, Epizyme, AstraZeneca

Educational grants: Roche

373 |PRELIMINARY CLINICAL DATA FROM A PHASE 1B

STUDY OF MAVORIXAFOR AND IBRUTINIB IN PATIENTS WITH

WALDENSTRÖM MACROGLOBULINEMIA WITH MYD88 AND

CXCR4 MUTATIONS

S. Treon

, C. Buske

, S. Thomas

, A. Branagan

, M. Dimopoulos

, J. J.

Castillo

, F. Garzon

, W. Tang

, R. Ronan

, S. Seyffert

, V. Garg

S. Ali

, A. Taveras

, A. Badarau

, K. Zmajkovicova

, S. Maier

B. Maierhofer

, J. Matous

Harvard Medical School, Bing Center for Waldenstrom's

Macroglobulinemia, Boston, USA,

University of Ulm, Comprehensive

Cancer Center and Institute of Experimental Cancer Research, Ulm,

Germany,

The University of Texas MD Anderson Cancer Center,

Department of Lymphoma/Myeloma, Division of Cancer Medicine,

Houston, USA,

Massachusetts General Hospital, Cancer Center, Boston,

USA,

University of Athens School of Medicine, Department of Clinical

Therapeutics, National and Kapodistrian, Athens, Greece,

Pharmaceuticals, Boston, USA,

Colorado Blood Cancer Institute, Sarah

Cannon Research Institute, Denver, USA

Introduction: Waldenström macroglobulinemia (WM) is a rare dis-

order of clonal proliferation of immunoglobulin M (IgM)‐secreting B

cells. Most patients with WM (>90%) have somatic mutations in

MYD88. Some (≤40%) also have WHIM‐like activating mutations in

CXCR4 (CXCR4

WHIM

) that impact response to Bruton tyrosine kinase

inhibitors (BTKi). C‐X‐C chemokine receptor type 4 (CXCR4) inhibi-

tion has been shown to sensitize CXCR4

WHIM

‐expressing cells to

ibrutinib. This study examines the safety and efﬁcacy of mavorixafor,

an oral CXCR4 antagonist, in combination with ibrutinib in patients

with WM with MYD88 and CXCR4 mutations.

Methods: This phase 1b, open‐label, single‐arm study

(NCT04274738) examines intrapatient dosing, safety, pharmacoki-

netics (PK), and pharmacodynamics (PD) of mavorixafor in combi-

nation with ibrutinib. Adults with clinicopathologic WM diagnosis,

consensus criteria indication for treatment, measurable disease, ≤3

prior therapies, and MYD88

L265P

and CXCR4

WHIM

mutations are

eligible. Patients are initiated on mavorixafor 200 mg and ibrutinib

420 mg, both oral and once‐daily (QD). Mavorixafor is escalated to

400 mg after 28 days if no dose‐limiting toxicities (DLTs) occur and

to 600 mg if 400 mg is deemed safe (<2/6 DLTs). Outcomes include

adverse events (AEs) and change from baseline in IgM, PK, and PD

(peripheral white blood cell [WBC] counts).

Results: At data cutoff in January 2021, 7 patients have enrolled and

remain on study. All 7 were escalated to 400 mg; 6 remain at 400 mg,

and 1 de‐escalated to 200 mg (median exposure, 90 days). None have

yet escalated to 600 mg. Fifty‐six AEs were observed (84% grade 1).

Of 50 AEs with complete assessment for causal relationship to study

drugs, 4 were deemed related to mavorixafor only, 7 to ibrutinib only,

and 18 to the combination. One patient experienced all AEs attributed

to mavorixafor only. Only 1 DLT was observed (grade 3 hypertension

deﬁnitively related to ibrutinib and possibly related to mavorixafor).

458

SUPPLEMENT ABSTRACTS

No AEs led to study discontinuation, and no serious AEs were

observed. Six of 7 patients showed IgM decrease after 1 cycle. Four

patients received ≥3 cycles; all had IgM decrease, with 2 achieving

≥50% decrease consistent with partial response (median decrease,

51.0%; range, 4.4%–84.5%). Mavorixafor and ibrutinib exposures

were consistent with previous single‐agent studies, and combination

treatment increased WBCs.

Conclusions: Our ﬁndings to date in patients with WM carrying

both MYD88 and CXCR4 mutations show that mavorixafor in com-

bination with ibrutinib is well tolerated at doses ≤400 mg QD. Un-

altered mavorixafor and ibrutinib exposures suggest no apparent

drug–drug interaction, and mavorixafor exposures tracked with

increased WBC counts. Combination of mavorixafor with ibrutinib

led to rapid and clinically meaningful decrease in IgM, suggesting

mavorixafor may sensitize CXCR4

WHIM

‐expressing cells to BTKi.

EA – previously submitted to EHA 2021.

The research was funded by: X4 Pharmaceuticals

Keywords: Molecular Targeted Therapies, Combination Therapies,

Indolent non‐Hodgkin lymphoma

Conﬂicts of interests pertinent to the abstract

S. Treon

Consultant or advisory role: PharmaCyclics/Abbvie, Janssen Phar-

maceuticals, Beigene

Research funding: PharmaCyclics/Abbvie, Janssen Pharmaceuticals,

Beigene, BMS, Eli Lilly

C. Buske

Consultant or advisory role: Roche, Janssen, Bayer, MSD, Celltrion

Honoraria: Roche, Janssen, Bayer, MSD, Celltrion

Research funding: Roche, Janssen, Bayer, MSD, Celltrion

S. Thomas

Consultant or advisory role: Beigene, Pharmacyclics, Bristol Myers

Squibb, Ascentage Phama, Genetech, X4 Pharmaceuticals,

A. Branagan

Consultant or advisory role: Beigene, Sanoﬁ‐Genzyme, Karyopharm,

Pharmacyclics

M. Dimopoulos

Honoraria: AMGEN, TAKEDA, JANSSEN, BMS, BEIGENE

J. J. Castillo

Consultant or advisory role: Abbvie, Beigene, Pharmacyclics

Research funding: Abbvie, Beigene, Pharmacyclics, TG Therapeutics

F. Garzon

Consultant or advisory role: X4 Pharmaceuticals

W. Tang

Employment or leadership position: X4 Pharmaceuticals

Stock ownership: X4 Pharmaceuticals

R. Ronan

Employment or leadership position: X4 Pharmaceuticals

S. Seyffert

Employment or leadership position: X4 Pharmaceuticals

Stock ownership: X4 Pharmaceuticals, PSU, ESPP

V. Garg

Employment or leadership position: X4 Pharmaceuticals

Stock ownership: X4 Pharmaceuticals

S. Ali

Employment or leadership position: X4 Pharmaceuticals

Stock ownership: X4 Pharmaceuticals

A. Taveras

Employment or leadership position: X4 Pharmaceuticals

A. Badarau

Employment or leadership position: X4 Pharmaceuticals

Stock ownership: X4 Pharmaceuticals

K. Zmajkovicova

Employment or leadership position: X4 Pharmaceuticals

Stock ownership: X4 Pharmaceuticals

Research funding: X4 Pharmaceuticals

S. Maier

Employment or leadership position: X4 Pharmaceuticals

Stock ownership: X4 Pharmaceuticals

Research funding: X4 Pharmaceuticals

B. Maierhofer

Employment or leadership position: X4 Pharmaceuticals

Stock ownership: X4 Pharmaceuticals

Research funding: X4 Pharmaceuticals

J. Matous

Consultant or advisory role: Janssen Pharmacyclics

374 |PHASE I STUDY OF THE CD19/CD3 HALF‐LIFE EXTENDED

BITE

MOLECULE AMG 562 IN RELAPSED/REFRACTORY DIFFUSE

LARGE B CELL LYMPHOMA, MANTLE CELL LYMPHOMA AND

FOLLICULAR LYMPHOMA

M. D. Mead

, L. L. Popplewell

, M. Subklewe

, A. Ghobadi

, J.

Kuruvilla

, A. Kimball

, C. Tuglus

, S. Agarwal

, J. Stieglmaier

Ronald Reagan UCLA Medical Center, Department of Hematology &

Oncology, Los Angeles, California, USA,

City of Hope National Medical

Center, Department of Hematology & Hematopoietic Cell Transplantation,

Duarte, USA,

LMU – University Hospital Munich, Department of

Hematology & Oncology, Munich, Germany,

Washington University

SUPPLEMENT ABSTRACTS

459

School of Medicine, Division of Oncology, St. Louis, USA,

Princess

Margaret Cancer Centre, University of Toronto, Division of Medical

Oncology and Hematology, Toronto, Canada,

Amgen Inc., Oncology TA,

Clinical Research, Thousand Oaks, USA,

Amgen Inc., Global Biostatistical

Science, Thousand Oaks, US,

Amgen Inc., Clinical Pharmacology,

Modeling and Simulation, South San Francisco, USA,

Amgen Research

(Munich) GmbH, Early Development Hematology/Oncology, Munich,

Germany

Introduction: AMG 562, a half‐life extended (HLE), BiTE

(bi‐

speciﬁc T cell engager) molecule, simultaneously engages CD19

on B cells and CD3 on T cells triggering T cell‐mediated serial

lysis of CD19‐expressing B cells. The results of a Phase I study

of AMG 562 in adult patients (pts) with relapsed or refractory

(R/R) diffuse large B cell lymphoma (DLBCL), mantle cell lym-

phoma (MCL), or follicular lymphoma (FL) are presented. The

study has since been terminated and the ﬁnal results are pre-

sented here.

Methods: Safety (primary), efﬁcacy, and pharmacokinetic (PK)

(secondary) outcomes following AMG 562 administration (IV, q1w

until disease progression, intolerance, or withdrawal of consent)

were evaluated in adults with biopsy‐proven DLBCL, FL, or MCL

(NCT03571828). This Phase I study was designed with an

initial dose exploration phase to be followed by a dose expansion

phase with the safety proﬁle informing the dose for the next

cohort.

Results: At data cutoff (18 August 2020), 9 pts (DLBCL, n =4; MCL,

n=2; FL, n =3; median age, 65 [56–73] years; median prior lines of

therapy, 4 [1–15]) had received ≥1 dose of AMG 562 at one of two

dose levels (DL; 0.1mg, n =8; 0.3 mg, n =1).

Treatment‐emergent adverse events (TEAEs) were observed in

all 9 pts and led to treatment (tx) discontinuation in 2 (both at 0.1

μg DL; pseudomonal sepsis [n =1, Gr 4], encephalopathy [n =1, Gr

4]). The encephalopathy AE was deemed to be a dose‐limiting

toxicity; subsequently, all subjects were premedicated with dexa-

methasone before each AMG 562 dose. No tx‐related fatalities

were observed.

For most post‐dose time points, AMG 562 serum concentrations

were below the limit of quantiﬁcation, limiting the estimation of PK

parameters. In the pt who developed encephalopathy, the end of

infusion concentration was 100‐fold higher relative to levels

observed in other pts.

Seven pts had progressive disease at data cutoff; response was

not assessed in 1 pt who discontinued tx. A complete metabolic

response (CMR; 2014 Lugano Classiﬁcation) was noted in a 73‐year‐

old female pt with relapsed DLBCL (0.1mg DL; 5 previous lines of

therapy; ECOG status 0). CMR was characterized by disappearance

of the target lesion with no residual mass (Lugano 5‐point scale

score: 1) and the absence of ﬂuorodeoxyglucose (FDG)‐avid focal

lesions in the bone marrow. CMR was observed by the end of cycle 1

and was maintained until the last FDG‐PET/CT scan before data

cutoff (cycle 6).

Conclusions: As the trial was terminated early, the limited dataset

precludes the drawing of any conclusions regarding the safety

(maximum tolerated dose/recommended Phase 2 dose) or efﬁcacy of

AMG 562. Nevertheless, the study provides proof‐of‐concept sup-

port for use of the HLE BiTE immuno‐oncology platform in a variety

of hematological malignancies.

The research was funded by: Amgen Inc.

Keywords: Immunotherapy

Conﬂicts of interests pertinent to the abstract

L. L. Popplewell

Educational grants: Novartis

Other remuneration: Pﬁzer, F. Hoffmann‐La Roche AG

M. Subklewe

Consultant or advisory role: Amgen, BMS/Celgene, Gilead, Janssen,

Novartis, Pﬁzer, Seattle Genetics

Research funding: Amgen, Gilead, Miltenyi, Morphosys, Roche,

Seattle Genetics

Other remuneration: Amgen, BMS/Celgene, Gilead, Pﬁzer

A. Ghobadi

Consultant or advisory role: Amgen, Celgene/BMS, Kite/Gilead,

Wugen

Research funding: Kite/Gilead, Amgen

J. Kuruvilla

Consultant or advisory role: Abbvie, BMS, Gilead, Karyopharm,

Merck, Roche, Seattle Genetics

Honoraria: Amgen, Antengene, Astra Zeneca, BMS, Gilead, Incyte,

Janssen, Karyopharm, Merck, Novartis, Pﬁzer, Roche, Seattle Ge-

netics, TG Therapeutics

Research funding: Canadian Cancer Society, Leukemia and Lym-

phoma Society Canada, Princess Margaret Cancer Foundation,

Janssen, Roche, Astra Zeneca

Other remuneration: Karyopharm (DSMB)

A. Kimball

Employment or leadership position: Amgen Inc.

Stock ownership: Amgen Inc.

C. Tuglus

Employment or leadership position: Amgen Inc.

Stock ownership: Amgen Inc.

S. Agarwal

Employment or leadership position: Amgen Inc.

Stock ownership: Amgen Inc.

J. Stieglmaier

Employment or leadership position: Amgen Research (Munich) GmbH

Stock ownership: Amgen Inc.

460

SUPPLEMENT ABSTRACTS

375 |PHASE I STUDY OF BENDAMUSTINE, RITUXIMAB,

IBRUTINIB, AND VENETOCLAX IN RELAPSED, REFRACTORY

MANTLE CELL LYMPHOMA

A. Kumar

, C. Batlevi

, P. Drullinsky

, C. Grieve

, L. Laraque

, A.

Joseph

, N. Mahajan

, M. Matasar

, D. Straus

, A. Zelenetz

, P.

Hamlin

Memorial Sloan Kettering Cancer Center, Medicine, New York, New York,

USA

Introduction: Bendamustine (B) and rituximab (R) with ibrutinib (IBR)

was well‐tolerated and efﬁcacious (Maddocks Blood 2015). Ibrutinib

and venetoclax (VEN) were synergistic and active in relapsed, re-

fractory (R/R) MCL (Tam NEJM 2018). VEN increases apoptotic

priming and can act as a chemosensitizing agent. The aim of this phase I

study was to deﬁne the safety and tolerability of VEN with BR‐IBR in

R/R MCL (Clinical trial: NCT03295240).

Methods: Patients (pts) with R/R MCL received six 28‐day cycles of

BR‐IBR‐VEN in a 3+3 dose escalation design. Dose level 1 (DL1)

included B 90 mg/m2 day (d) 1 and 2, R 375 mg/m2 d1, and IBR

560mg daily. For cycle 1, VEN dose ramp‐up was 20, 50, 100, and

200mg daily weekly. During cycles 2‐6, VEN 400mg daily was given

d1‐5. Dosing cohorts with longer duration of VEN (7d and 10d) were

planned. The study was amended to include dose level ‐1 (DL‐1) with

B at 70mg/m2, d1 and 2.

Results: As of March 2021, 7 pts were treated with BR‐IBR‐VEN

at DL1 and 3 pts at DL‐1. Baseline characteristics were median age

68 years (range 60 ‐81); 90% male; 80% with Ki67 ≥30%; 10% blastic

MCL; 40% TP53 mutant, and 100% one prior line of treatment (tx).

One dose limiting toxicity (DLT) was observed at DL1: grade 3

thrombocytopenia lasting >14 days. At DL1, 2 pts had only 5 cycles

of BR‐IBR‐VEN due to cytopenias (grade 3 neutropenia and grade 2

thrombocytopenia) despite B dose reduction. The frequent tx‐related

adverse events (AEs) were thrombocytopenia (n =7), constipation

(n =6), fatigue (n =4), neutropenia (n =3), and nausea (n =4). The

grade 3‐4 AEs were neutropenia (n =1) thrombocytopenia (n =3),

and rash (n =1). There was one tx‐related serious AE: an elderly pt

was brieﬂy hospitalized during VEN ramp‐up for volume overload

and diuretic tx. Most pts had count recovery post‐tx, but one had

persistent neutropenia. Given cytopenias observed at DL1, the study

was amended to include DL‐1. At DL‐1 (n =3), there were no DLTs,

dose reductions or delays, signiﬁcant cytopenias, or ≥grade 3 tx‐

related AEs.

The overall response rate (ORR) was 80% (8/10), all com-

plete responses (CRs). For TP53 mutant MCL, the ORR was 50%

(2/4), all CRs. There were 5 deaths (Fig 1). Two TP53 mutant

MCL pts with primary refractory disease died of disease rapidly.

A 17p deleted MCL pt achieved a CR and underwent allogeneic

stem cell transplant and died of pneumonia (PNA) 10 mos after.

A pt in CR with post‐tx neutropenia died of PNA. The TP53

mutant MCL pt with a DLT after 2 cycles of BR‐IBR‐VEN

received single‐agent BTKi for 27 mos until progression and

then died of COVID.

Conclusions: BR‐IBR‐VEN showed acceptable safety and tolerability

and preliminary efﬁcacy at DL‐1. During the COVID pandemic,

enrollment for this immunosuppressive 4‐drug regimen with intrave-

nous tx was challenging. The study was closed after enrolling 3 of 6 pts

at DL‐1 and a deﬁnitive maximum tolerated dose was not determined.

Future study is needed to determine the role of BR‐IBR‐VEN in MCL.

The research was funded by: Genentech

Keywords: Lymphoid Cancers – Other, Chemotherapy, Molecular

Targeted Therapies

Conﬂicts of interests pertinent to the abstract

A. Kumar

Consultant or advisory role: Kite Pharmaceuticals, Astra Zeneca, and

Celgene

FIGURE 1 Swimmer Plot for BR‐VEN

treated patients (n =10)

SUPPLEMENT ABSTRACTS

461